Merge branch 'pm-opp' into pm-cpufreq

Documentation/devicetree/bindings/dma/apm-xgene-dma.txt

@@ -35,7 +35,7 @@ Example:
 			device_type = "dma";
 			reg = <0x0 0x1f270000 0x0 0x10000>,
 			      <0x0 0x1f200000 0x0 0x10000>,
-			      <0x0 0x1b008000 0x0 0x2000>,
+			      <0x0 0x1b000000 0x0 0x400000>,
 			      <0x0 0x1054a000 0x0 0x100>;
 			interrupts = <0x0 0x82 0x4>,
 				     <0x0 0xb8 0x4>,

Documentation/devicetree/bindings/opp/opp.txt

@@ -88,7 +88,7 @@ This defines voltage-current-frequency combinations along with other related
 properties.
 
 Required properties:
-- opp-hz: Frequency in Hz
+- opp-hz: Frequency in Hz, expressed as a 64-bit big-endian integer.
 
 Optional properties:
 - opp-microvolt: voltage in micro Volts.
@@ -158,20 +158,20 @@ Example 1: Single cluster Dual-core ARM cortex A9, switch DVFS states together.
 		opp-shared;
 
 		opp00 {
-			opp-hz = <1000000000>;
+			opp-hz = /bits/ 64 <1000000000>;
 			opp-microvolt = <970000 975000 985000>;
 			opp-microamp = <70000>;
 			clock-latency-ns = <300000>;
 			opp-suspend;
 		};
 		opp01 {
-			opp-hz = <1100000000>;
+			opp-hz = /bits/ 64 <1100000000>;
 			opp-microvolt = <980000 1000000 1010000>;
 			opp-microamp = <80000>;
 			clock-latency-ns = <310000>;
 		};
 		opp02 {
-			opp-hz = <1200000000>;
+			opp-hz = /bits/ 64 <1200000000>;
 			opp-microvolt = <1025000>;
 			clock-latency-ns = <290000>;
 			turbo-mode;
@@ -237,20 +237,20 @@ independently.
 		 */
 
 		opp00 {
-			opp-hz = <1000000000>;
+			opp-hz = /bits/ 64 <1000000000>;
 			opp-microvolt = <970000 975000 985000>;
 			opp-microamp = <70000>;
 			clock-latency-ns = <300000>;
 			opp-suspend;
 		};
 		opp01 {
-			opp-hz = <1100000000>;
+			opp-hz = /bits/ 64 <1100000000>;
 			opp-microvolt = <980000 1000000 1010000>;
 			opp-microamp = <80000>;
 			clock-latency-ns = <310000>;
 		};
 		opp02 {
-			opp-hz = <1200000000>;
+			opp-hz = /bits/ 64 <1200000000>;
 			opp-microvolt = <1025000>;
 			opp-microamp = <90000;
 			lock-latency-ns = <290000>;
@@ -313,20 +313,20 @@ DVFS state together.
 		opp-shared;
 
 		opp00 {
-			opp-hz = <1000000000>;
+			opp-hz = /bits/ 64 <1000000000>;
 			opp-microvolt = <970000 975000 985000>;
 			opp-microamp = <70000>;
 			clock-latency-ns = <300000>;
 			opp-suspend;
 		};
 		opp01 {
-			opp-hz = <1100000000>;
+			opp-hz = /bits/ 64 <1100000000>;
 			opp-microvolt = <980000 1000000 1010000>;
 			opp-microamp = <80000>;
 			clock-latency-ns = <310000>;
 		};
 		opp02 {
-			opp-hz = <1200000000>;
+			opp-hz = /bits/ 64 <1200000000>;
 			opp-microvolt = <1025000>;
 			opp-microamp = <90000>;
 			clock-latency-ns = <290000>;
@@ -339,20 +339,20 @@ DVFS state together.
 		opp-shared;
 
 		opp10 {
-			opp-hz = <1300000000>;
+			opp-hz = /bits/ 64 <1300000000>;
 			opp-microvolt = <1045000 1050000 1055000>;
 			opp-microamp = <95000>;
 			clock-latency-ns = <400000>;
 			opp-suspend;
 		};
 		opp11 {
-			opp-hz = <1400000000>;
+			opp-hz = /bits/ 64 <1400000000>;
 			opp-microvolt = <1075000>;
 			opp-microamp = <100000>;
 			clock-latency-ns = <400000>;
 		};
 		opp12 {
-			opp-hz = <1500000000>;
+			opp-hz = /bits/ 64 <1500000000>;
 			opp-microvolt = <1010000 1100000 1110000>;
 			opp-microamp = <95000>;
 			clock-latency-ns = <400000>;
@@ -379,7 +379,7 @@ Example 4: Handling multiple regulators
 		opp-shared;
 
 		opp00 {
-			opp-hz = <1000000000>;
+			opp-hz = /bits/ 64 <1000000000>;
 			opp-microvolt = <970000>, /* Supply 0 */
 					<960000>, /* Supply 1 */
 					<960000>; /* Supply 2 */
@@ -392,7 +392,7 @@ Example 4: Handling multiple regulators
 		/* OR */
 
 		opp00 {
-			opp-hz = <1000000000>;
+			opp-hz = /bits/ 64 <1000000000>;
 			opp-microvolt = <970000 975000 985000>, /* Supply 0 */
 					<960000 965000 975000>, /* Supply 1 */
 					<960000 965000 975000>; /* Supply 2 */
@@ -405,7 +405,7 @@ Example 4: Handling multiple regulators
 		/* OR */
 
 		opp00 {
-			opp-hz = <1000000000>;
+			opp-hz = /bits/ 64 <1000000000>;
 			opp-microvolt = <970000 975000 985000>, /* Supply 0 */
 					<960000 965000 975000>, /* Supply 1 */
 					<960000 965000 975000>; /* Supply 2 */
@@ -437,12 +437,12 @@ Example 5: Multiple OPP tables
 		opp-shared;
 
 		opp00 {
-			opp-hz = <600000000>;
+			opp-hz = /bits/ 64 <600000000>;
 			...
 		};
 
 		opp01 {
-			opp-hz = <800000000>;
+			opp-hz = /bits/ 64 <800000000>;
 			...
 		};
 	};
@@ -453,12 +453,12 @@ Example 5: Multiple OPP tables
 		opp-shared;
 
 		opp10 {
-			opp-hz = <1000000000>;
+			opp-hz = /bits/ 64 <1000000000>;
 			...
 		};
 
 		opp11 {
-			opp-hz = <1100000000>;
+			opp-hz = /bits/ 64 <1100000000>;
 			...
 		};
 	};

Documentation/devicetree/bindings/sound/mt8173-max98090.txt

@@ -3,11 +3,13 @@ MT8173 with MAX98090 CODEC
 Required properties:
 - compatible : "mediatek,mt8173-max98090"
 - mediatek,audio-codec: the phandle of the MAX98090 audio codec
+- mediatek,platform: the phandle of MT8173 ASoC platform
 
 Example:
 
 	sound {
 		compatible = "mediatek,mt8173-max98090";
 		mediatek,audio-codec = <&max98090>;
+		mediatek,platform = <&afe>;
 	};
 

Documentation/devicetree/bindings/sound/mt8173-rt5650-rt5676.txt

@@ -3,11 +3,13 @@ MT8173 with RT5650 RT5676 CODECS
 Required properties:
 - compatible : "mediatek,mt8173-rt5650-rt5676"
 - mediatek,audio-codec: the phandles of rt5650 and rt5676 codecs
+- mediatek,platform: the phandle of MT8173 ASoC platform
 
 Example:
 
 	sound {
 		compatible = "mediatek,mt8173-rt5650-rt5676";
 		mediatek,audio-codec = <&rt5650 &rt5676>;
+		mediatek,platform = <&afe>;
 	};
 

Documentation/devicetree/bindings/spi/spi-ath79.txt

@@ -3,7 +3,7 @@ Binding for Qualcomm Atheros AR7xxx/AR9xxx SPI controller
 Required properties:
 - compatible: has to be "qca,<soc-type>-spi", "qca,ar7100-spi" as fallback.
 - reg: Base address and size of the controllers memory area
-- clocks: phandle to the AHB clock.
+- clocks: phandle of the AHB clock.
 - clock-names: has to be "ahb".
 - #address-cells: <1>, as required by generic SPI binding.
 - #size-cells: <0>, also as required by generic SPI binding.
@@ -12,9 +12,9 @@ Child nodes as per the generic SPI binding.
 
 Example:
 
-	spi@1F000000 {
+	spi@1f000000 {
 		compatible = "qca,ar9132-spi", "qca,ar7100-spi";
-		reg = <0x1F000000 0x10>;
+		reg = <0x1f000000 0x10>;
 
 		clocks = <&pll 2>;
 		clock-names = "ahb";

Documentation/hwmon/nct7904

@@ -35,11 +35,11 @@ temp1_input		Local temperature (1/1000 degree,
 temp[2-9]_input		CPU temperatures (1/1000 degree,
 			0.125 degree resolution)
 
-fan[1-4]_mode		R/W, 0/1 for manual or SmartFan mode
+pwm[1-4]_enable		R/W, 1/2 for manual or SmartFan mode
 			Setting SmartFan mode is supported only if it has been
 			previously configured by BIOS (or configuration EEPROM)
 
-fan[1-4]_pwm		R/O in SmartFan mode, R/W in manual control mode
+pwm[1-4]		R/O in SmartFan mode, R/W in manual control mode
 
 The driver checks sensor control registers and does not export the sensors
 that are not enabled. Anyway, a sensor that is enabled may actually be not

Documentation/target/tcm_mod_builder.py

@@ -199,7 +199,8 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name):
 	buf += "#include <linux/string.h>\n"
 	buf += "#include <linux/configfs.h>\n"
 	buf += "#include <linux/ctype.h>\n"
-	buf += "#include <asm/unaligned.h>\n\n"
+	buf += "#include <asm/unaligned.h>\n"
+	buf += "#include <scsi/scsi_proto.h>\n\n"
 	buf += "#include <target/target_core_base.h>\n"
 	buf += "#include <target/target_core_fabric.h>\n"
 	buf += "#include <target/target_core_fabric_configfs.h>\n"
@@ -230,8 +231,14 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name):
 	buf += "	}\n"
 	buf += "	tpg->" + fabric_mod_port + " = " + fabric_mod_port + ";\n"
 	buf += "	tpg->" + fabric_mod_port + "_tpgt = tpgt;\n\n"
-	buf += "	ret = core_tpg_register(&" + fabric_mod_name + "_ops, wwn,\n"
+
-	buf += "				&tpg->se_tpg, SCSI_PROTOCOL_SAS);\n"
+	if proto_ident == "FC":
+		buf += "	ret = core_tpg_register(wwn, &tpg->se_tpg, SCSI_PROTOCOL_FCP);\n"
+	elif proto_ident == "SAS":
+		buf += "	ret = core_tpg_register(wwn, &tpg->se_tpg, SCSI_PROTOCOL_SAS);\n"
+	elif proto_ident == "iSCSI":
+		buf += "	ret = core_tpg_register(wwn, &tpg->se_tpg, SCSI_PROTOCOL_ISCSI);\n"
+
 	buf += "	if (ret < 0) {\n"
 	buf += "		kfree(tpg);\n"
 	buf += "		return NULL;\n"
@@ -292,7 +299,7 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name):
 
 	buf += "static const struct target_core_fabric_ops " + fabric_mod_name + "_ops = {\n"
 	buf += "	.module				= THIS_MODULE,\n"
-	buf += "	.name				= " + fabric_mod_name + ",\n"
+	buf += "	.name				= \"" + fabric_mod_name + "\",\n"
 	buf += "	.get_fabric_name		= " + fabric_mod_name + "_get_fabric_name,\n"
 	buf += "	.tpg_get_wwn			= " + fabric_mod_name + "_get_fabric_wwn,\n"
 	buf += "	.tpg_get_tag			= " + fabric_mod_name + "_get_tag,\n"
@@ -322,17 +329,17 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name):
 	buf += "	.fabric_make_tpg		= " + fabric_mod_name + "_make_tpg,\n"
 	buf += "	.fabric_drop_tpg		= " + fabric_mod_name + "_drop_tpg,\n"
 	buf += "\n"
-	buf += "	.tfc_wwn_attrs			= " + fabric_mod_name + "_wwn_attrs;\n"
+	buf += "	.tfc_wwn_attrs			= " + fabric_mod_name + "_wwn_attrs,\n"
 	buf += "};\n\n"
 
 	buf += "static int __init " + fabric_mod_name + "_init(void)\n"
 	buf += "{\n"
-	buf += "	return target_register_template(" + fabric_mod_name + "_ops);\n"
+	buf += "	return target_register_template(&" + fabric_mod_name + "_ops);\n"
 	buf += "};\n\n"
 
 	buf += "static void __exit " + fabric_mod_name + "_exit(void)\n"
 	buf += "{\n"
-	buf += "	target_unregister_template(" + fabric_mod_name + "_ops);\n"
+	buf += "	target_unregister_template(&" + fabric_mod_name + "_ops);\n"
 	buf += "};\n\n"
 
 	buf += "MODULE_DESCRIPTION(\"" + fabric_mod_name.upper() + " series fabric driver\");\n"

MAINTAINERS

@@ -5600,6 +5600,7 @@ F:	kernel/irq/
 IRQCHIP DRIVERS
 M:	Thomas Gleixner <tglx@linutronix.de>
 M:	Jason Cooper <jason@lakedaemon.net>
+M:	Marc Zyngier <marc.zyngier@arm.com>
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core
@@ -5608,11 +5609,14 @@ F:	Documentation/devicetree/bindings/interrupt-controller/
 F:	drivers/irqchip/
 
 IRQ DOMAINS (IRQ NUMBER MAPPING LIBRARY)
-M:	Benjamin Herrenschmidt <benh@kernel.crashing.org>
+M:	Jiang Liu <jiang.liu@linux.intel.com>
+M:	Marc Zyngier <marc.zyngier@arm.com>
 S:	Maintained
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core
 F:	Documentation/IRQ-domain.txt
 F:	include/linux/irqdomain.h
 F:	kernel/irq/irqdomain.c
+F:	kernel/irq/msi.c
 
 ISAPNP
 M:	Jaroslav Kysela <perex@perex.cz>

Makefile

@@ -1,7 +1,7 @@
 VERSION = 4
 PATCHLEVEL = 2
 SUBLEVEL = 0
-EXTRAVERSION = -rc4
+EXTRAVERSION = -rc5
 NAME = Hurr durr I'ma sheep
 
 # *DOCUMENTATION*

arch/arm/boot/dts/exynos3250.dtsi

@@ -138,8 +138,8 @@
 
 		mipi_phy: video-phy@10020710 {
 			compatible = "samsung,s5pv210-mipi-video-phy";
-			reg = <0x10020710 8>;
 			#phy-cells = <1>;
+			syscon = <&pmu_system_controller>;
 		};
 
 		pd_cam: cam-power-domain@10023C00 {

arch/arm/boot/dts/exynos4210-origen.dts

@@ -127,6 +127,10 @@
 	};
 };
 
+&cpu0 {
+	cpu0-supply = <&buck1_reg>;
+};
+
 &fimd {
 	pinctrl-0 = <&lcd_en &lcd_clk &lcd_data24 &pwm0_out>;
 	pinctrl-names = "default";

arch/arm/boot/dts/exynos4210-trats.dts

@@ -188,6 +188,10 @@
 	};
 };
 
+&cpu0 {
+	cpu0-supply = <&varm_breg>;
+};
+
 &dsi_0 {
 	vddcore-supply = <&vusb_reg>;
 	vddio-supply = <&vmipi_reg>;

arch/arm/boot/dts/exynos4210-universal_c210.dts

@@ -548,6 +548,10 @@
 	};
 };
 
+&cpu0 {
+	cpu0-supply = <&vdd_arm_reg>;
+};
+
 &pinctrl_1 {
 	hdmi_hpd: hdmi-hpd {
 		samsung,pins = "gpx3-7";

arch/arm/boot/dts/exynos4210.dtsi

@@ -40,6 +40,18 @@
 			device_type = "cpu";
 			compatible = "arm,cortex-a9";
 			reg = <0x900>;
+			clocks = <&clock CLK_ARM_CLK>;
+			clock-names = "cpu";
+			clock-latency = <160000>;
+
+			operating-points = <
+				1200000 1250000
+				1000000 1150000
+				800000	1075000
+				500000	975000
+				400000	975000
+				200000	950000
+			>;
 			cooling-min-level = <4>;
 			cooling-max-level = <2>;
 			#cooling-cells = <2>; /* min followed by max */

arch/arm/boot/dts/imx35.dtsi

@@ -286,8 +286,8 @@
 			can1: can@53fe4000 {
 				compatible = "fsl,imx35-flexcan", "fsl,p1010-flexcan";
 				reg = <0x53fe4000 0x1000>;
-				clocks = <&clks 33>;
+				clocks = <&clks 33>, <&clks 33>;
-				clock-names = "ipg";
+				clock-names = "ipg", "per";
 				interrupts = <43>;
 				status = "disabled";
 			};
@@ -295,8 +295,8 @@
 			can2: can@53fe8000 {
 				compatible = "fsl,imx35-flexcan", "fsl,p1010-flexcan";
 				reg = <0x53fe8000 0x1000>;
-				clocks = <&clks 34>;
+				clocks = <&clks 34>, <&clks 34>;
-				clock-names = "ipg";
+				clock-names = "ipg", "per";
 				interrupts = <44>;
 				status = "disabled";
 			};

arch/arm/boot/dts/k2e-clocks.dtsi

@@ -13,9 +13,8 @@ clocks {
 		#clock-cells = <0>;
 		compatible = "ti,keystone,main-pll-clock";
 		clocks = <&refclksys>;
-		reg = <0x02620350 4>, <0x02310110 4>;
+		reg = <0x02620350 4>, <0x02310110 4>, <0x02310108 4>;
-		reg-names = "control", "multiplier";
+		reg-names = "control", "multiplier", "post-divider";
-		fixed-postdiv = <2>;
 	};
 
 	papllclk: papllclk@2620358 {

arch/arm/boot/dts/k2hk-clocks.dtsi

@@ -22,9 +22,8 @@ clocks {
 		#clock-cells = <0>;
 		compatible = "ti,keystone,main-pll-clock";
 		clocks = <&refclksys>;
-		reg = <0x02620350 4>, <0x02310110 4>;
+		reg = <0x02620350 4>, <0x02310110 4>, <0x02310108 4>;
-		reg-names = "control", "multiplier";
+		reg-names = "control", "multiplier", "post-divider";
-		fixed-postdiv = <2>;
 	};
 
 	papllclk: papllclk@2620358 {

arch/arm/boot/dts/k2l-clocks.dtsi

@@ -22,9 +22,8 @@ clocks {
 		#clock-cells = <0>;
 		compatible = "ti,keystone,main-pll-clock";
 		clocks = <&refclksys>;
-		reg = <0x02620350 4>, <0x02310110 4>;
+		reg = <0x02620350 4>, <0x02310110 4>, <0x02310108 4>;
-		reg-names = "control", "multiplier";
+		reg-names = "control", "multiplier", "post-divider";
-		fixed-postdiv = <2>;
 	};
 
 	papllclk: papllclk@2620358 {

arch/arm/boot/dts/ste-nomadik-nhk15.dts

@@ -17,6 +17,7 @@
 	};
 
 	aliases {
+		serial1 = &uart1;
 		stmpe-i2c0 = &stmpe0;
 		stmpe-i2c1 = &stmpe1;
 	};

arch/arm/boot/dts/ste-nomadik-s8815.dts

@@ -15,6 +15,10 @@
 		bootargs = "root=/dev/ram0 console=ttyAMA1,115200n8 earlyprintk";
 	};
 
+	aliases {
+		serial1 = &uart1;
+	};
+
 	src@101e0000 {
 		/* These chrystal drivers are not used on this board */
 		disable-sxtalo;

arch/arm/boot/dts/ste-nomadik-stn8815.dtsi

@@ -757,6 +757,7 @@
 			clock-names = "uartclk", "apb_pclk";
 			pinctrl-names = "default";
 			pinctrl-0 = <&uart0_default_mux>;
+			status = "disabled";
 		};
 
 		uart1: uart@101fb000 {

arch/arm/mach-omap2/omap_hwmod.c

@@ -2373,6 +2373,9 @@ static int of_dev_hwmod_lookup(struct device_node *np,
  * registers.  This address is needed early so the OCP registers that
  * are part of the device's address space can be ioremapped properly.
  *
+ * If SYSC access is not needed, the registers will not be remapped
+ * and non-availability of MPU access is not treated as an error.
+ *
  * Returns 0 on success, -EINVAL if an invalid hwmod is passed, and
  * -ENXIO on absent or invalid register target address space.
  */
@@ -2387,6 +2390,11 @@ static int __init _init_mpu_rt_base(struct omap_hwmod *oh, void *data,
 
 	_save_mpu_port_index(oh);
 
+	/* if we don't need sysc access we don't need to ioremap */
+	if (!oh->class->sysc)
+		return 0;
+
+	/* we can't continue without MPU PORT if we need sysc access */
 	if (oh->_int_flags & _HWMOD_NO_MPU_PORT)
 		return -ENXIO;
 
@@ -2396,8 +2404,10 @@ static int __init _init_mpu_rt_base(struct omap_hwmod *oh, void *data,
 			 oh->name);
 
 		/* Extract the IO space from device tree blob */
-		if (!np)
+		if (!np) {
+			pr_err("omap_hwmod: %s: no dt node\n", oh->name);
 			return -ENXIO;
+		}
 
 		va_start = of_iomap(np, index + oh->mpu_rt_idx);
 	} else {
@@ -2456,13 +2466,11 @@ static int __init _init(struct omap_hwmod *oh, void *data)
 				oh->name, np->name);
 	}
 
-	if (oh->class->sysc) {
+	r = _init_mpu_rt_base(oh, NULL, index, np);
-		r = _init_mpu_rt_base(oh, NULL, index, np);
+	if (r < 0) {
-		if (r < 0) {
+		WARN(1, "omap_hwmod: %s: doesn't have mpu register target base\n",
-			WARN(1, "omap_hwmod: %s: doesn't have mpu register target base\n",
+		     oh->name);
-			     oh->name);
+		return 0;
-			return 0;
-		}
 	}
 
 	r = _init_clocks(oh, NULL);

arch/arm/mach-omap2/omap_hwmod_7xx_data.c

@@ -827,8 +827,7 @@ static struct omap_hwmod_class_sysconfig dra7xx_gpmc_sysc = {
 	.syss_offs	= 0x0014,
 	.sysc_flags	= (SYSC_HAS_AUTOIDLE | SYSC_HAS_SIDLEMODE |
 			   SYSC_HAS_SOFTRESET | SYSS_HAS_RESET_STATUS),
-	.idlemodes	= (SIDLE_FORCE | SIDLE_NO | SIDLE_SMART |
+	.idlemodes	= (SIDLE_FORCE | SIDLE_NO | SIDLE_SMART),
-			   SIDLE_SMART_WKUP),
 	.sysc_fields	= &omap_hwmod_sysc_type1,
 };
 
@@ -844,7 +843,7 @@ static struct omap_hwmod dra7xx_gpmc_hwmod = {
 	.class		= &dra7xx_gpmc_hwmod_class,
 	.clkdm_name	= "l3main1_clkdm",
 	/* Skip reset for CONFIG_OMAP_GPMC_DEBUG for bootloader timings */
-	.flags		= HWMOD_SWSUP_SIDLE | DEBUG_OMAP_GPMC_HWMOD_FLAGS,
+	.flags		= DEBUG_OMAP_GPMC_HWMOD_FLAGS,
 	.main_clk	= "l3_iclk_div",
 	.prcm = {
 		.omap4 = {

arch/arm64/boot/dts/apm/apm-storm.dtsi

@@ -823,7 +823,7 @@
 			device_type = "dma";
 			reg = <0x0 0x1f270000 0x0 0x10000>,
 			      <0x0 0x1f200000 0x0 0x10000>,
-			      <0x0 0x1b008000 0x0 0x2000>,
+			      <0x0 0x1b000000 0x0 0x400000>,
 			      <0x0 0x1054a000 0x0 0x100>;
 			interrupts = <0x0 0x82 0x4>,
 				     <0x0 0xb8 0x4>,

arch/arm64/kernel/efi.c

@@ -122,12 +122,12 @@ static int __init uefi_init(void)
 
 	/* Show what we know for posterity */
 	c16 = early_memremap(efi_to_phys(efi.systab->fw_vendor),
-			     sizeof(vendor));
+			     sizeof(vendor) * sizeof(efi_char16_t));
 	if (c16) {
 		for (i = 0; i < (int) sizeof(vendor) - 1 && *c16; ++i)
 			vendor[i] = c16[i];
 		vendor[i] = '\0';
-		early_memunmap(c16, sizeof(vendor));
+		early_memunmap(c16, sizeof(vendor) * sizeof(efi_char16_t));
 	}
 
 	pr_info("EFI v%u.%.02u by %s\n",

arch/avr32/mach-at32ap/clock.c

@@ -80,6 +80,9 @@ int clk_enable(struct clk *clk)
 {
 	unsigned long flags;
 
+	if (!clk)
+		return 0;
+
 	spin_lock_irqsave(&clk_lock, flags);
 	__clk_enable(clk);
 	spin_unlock_irqrestore(&clk_lock, flags);
@@ -106,6 +109,9 @@ void clk_disable(struct clk *clk)
 {
 	unsigned long flags;
 
+	if (IS_ERR_OR_NULL(clk))
+		return;
+
 	spin_lock_irqsave(&clk_lock, flags);
 	__clk_disable(clk);
 	spin_unlock_irqrestore(&clk_lock, flags);
@@ -117,6 +123,9 @@ unsigned long clk_get_rate(struct clk *clk)
 	unsigned long flags;
 	unsigned long rate;
 
+	if (!clk)
+		return 0;
+
 	spin_lock_irqsave(&clk_lock, flags);
 	rate = clk->get_rate(clk);
 	spin_unlock_irqrestore(&clk_lock, flags);
@@ -129,6 +138,9 @@ long clk_round_rate(struct clk *clk, unsigned long rate)
 {
 	unsigned long flags, actual_rate;
 
+	if (!clk)
+		return 0;
+
 	if (!clk->set_rate)
 		return -ENOSYS;
 
@@ -145,6 +157,9 @@ int clk_set_rate(struct clk *clk, unsigned long rate)
 	unsigned long flags;
 	long ret;
 
+	if (!clk)
+		return 0;
+
 	if (!clk->set_rate)
 		return -ENOSYS;
 
@@ -161,6 +176,9 @@ int clk_set_parent(struct clk *clk, struct clk *parent)
 	unsigned long flags;
 	int ret;
 
+	if (!clk)
+		return 0;
+
 	if (!clk->set_parent)
 		return -ENOSYS;
 
@@ -174,7 +192,7 @@ EXPORT_SYMBOL(clk_set_parent);
 
 struct clk *clk_get_parent(struct clk *clk)
 {
-	return clk->parent;
+	return !clk ? NULL : clk->parent;
 }
 EXPORT_SYMBOL(clk_get_parent);
 

arch/powerpc/platforms/powernv/eeh-powernv.c

@@ -1478,7 +1478,7 @@ static int pnv_eeh_next_error(struct eeh_pe **pe)
 	}
 
 	/* Unmask the event */
-	if (eeh_enabled())
+	if (ret == EEH_NEXT_ERR_NONE && eeh_enabled())
 		enable_irq(eeh_event_irq);
 
 	return ret;

arch/powerpc/platforms/powernv/pci-ioda.c

@@ -2220,7 +2220,7 @@ static void pnv_pci_ioda_setup_opal_tce_kill(struct pnv_phb *phb)
 
 static __be64 *pnv_pci_ioda2_table_do_alloc_pages(int nid, unsigned shift,
 		unsigned levels, unsigned long limit,
-		unsigned long *current_offset)
+		unsigned long *current_offset, unsigned long *total_allocated)
 {
 	struct page *tce_mem = NULL;
 	__be64 *addr, *tmp;
@@ -2236,6 +2236,7 @@ static __be64 *pnv_pci_ioda2_table_do_alloc_pages(int nid, unsigned shift,
 	}
 	addr = page_address(tce_mem);
 	memset(addr, 0, allocated);
+	*total_allocated += allocated;
 
 	--levels;
 	if (!levels) {
@@ -2245,7 +2246,7 @@ static __be64 *pnv_pci_ioda2_table_do_alloc_pages(int nid, unsigned shift,
 
 	for (i = 0; i < entries; ++i) {
 		tmp = pnv_pci_ioda2_table_do_alloc_pages(nid, shift,
-				levels, limit, current_offset);
+				levels, limit, current_offset, total_allocated);
 		if (!tmp)
 			break;
 
@@ -2267,7 +2268,7 @@ static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset,
 		struct iommu_table *tbl)
 {
 	void *addr;
-	unsigned long offset = 0, level_shift;
+	unsigned long offset = 0, level_shift, total_allocated = 0;
 	const unsigned window_shift = ilog2(window_size);
 	unsigned entries_shift = window_shift - page_shift;
 	unsigned table_shift = max_t(unsigned, entries_shift + 3, PAGE_SHIFT);
@@ -2286,7 +2287,7 @@ static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset,
 
 	/* Allocate TCE table */
 	addr = pnv_pci_ioda2_table_do_alloc_pages(nid, level_shift,
-			levels, tce_table_size, &offset);
+			levels, tce_table_size, &offset, &total_allocated);
 
 	/* addr==NULL means that the first level allocation failed */
 	if (!addr)
@@ -2308,7 +2309,7 @@ static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset,
 			page_shift);
 	tbl->it_level_size = 1ULL << (level_shift - 3);
 	tbl->it_indirect_levels = levels - 1;
-	tbl->it_allocated_size = offset;
+	tbl->it_allocated_size = total_allocated;
 
 	pr_devel("Created TCE table: ws=%08llx ts=%lx @%08llx\n",
 			window_size, tce_table_size, bus_offset);

arch/s390/kernel/cache.c

@@ -138,6 +138,8 @@ int init_cache_level(unsigned int cpu)
 	union cache_topology ct;
 	enum cache_type ctype;
 
+	if (!test_facility(34))
+		return -EOPNOTSUPP;
 	if (!this_cpu_ci)
 		return -EINVAL;
 	ct.raw = ecag(EXTRACT_TOPOLOGY, 0, 0);

arch/s390/net/bpf_jit_comp.c

@@ -448,13 +448,13 @@ static void bpf_jit_prologue(struct bpf_jit *jit)
 		EMIT6_DISP_LH(0xe3000000, 0x0004, REG_SKB_DATA, REG_0,
 			      BPF_REG_1, offsetof(struct sk_buff, data));
 	}
-	/* BPF compatibility: clear A (%b7) and X (%b8) registers */
+	/* BPF compatibility: clear A (%b0) and X (%b7) registers */
-	if (REG_SEEN(BPF_REG_7))
+	if (REG_SEEN(BPF_REG_A))
-		/* lghi %b7,0 */
+		/* lghi %ba,0 */
-		EMIT4_IMM(0xa7090000, BPF_REG_7, 0);
+		EMIT4_IMM(0xa7090000, BPF_REG_A, 0);
-	if (REG_SEEN(BPF_REG_8))
+	if (REG_SEEN(BPF_REG_X))
-		/* lghi %b8,0 */
+		/* lghi %bx,0 */
-		EMIT4_IMM(0xa7090000, BPF_REG_8, 0);
+		EMIT4_IMM(0xa7090000, BPF_REG_X, 0);
 }
 
 /*

arch/x86/boot/compressed/eboot.c

@@ -1193,6 +1193,10 @@ static efi_status_t setup_e820(struct boot_params *params,
 		unsigned int e820_type = 0;
 		unsigned long m = efi->efi_memmap;
 
+#ifdef CONFIG_X86_64
+		m |= (u64)efi->efi_memmap_hi << 32;
+#endif
+
 		d = (efi_memory_desc_t *)(m + (i * efi->efi_memdesc_size));
 		switch (d->type) {
 		case EFI_RESERVED_TYPE:

arch/x86/include/asm/desc.h

@@ -280,21 +280,6 @@ static inline void clear_LDT(void)
 	set_ldt(NULL, 0);
 }
 
-/*
- * load one particular LDT into the current CPU
- */
-static inline void load_LDT_nolock(mm_context_t *pc)
-{
-	set_ldt(pc->ldt, pc->size);
-}
-
-static inline void load_LDT(mm_context_t *pc)
-{
-	preempt_disable();
-	load_LDT_nolock(pc);
-	preempt_enable();
-}
-
 static inline unsigned long get_desc_base(const struct desc_struct *desc)
 {
 	return (unsigned)(desc->base0 | ((desc->base1) << 16) | ((desc->base2) << 24));

arch/x86/include/asm/mmu.h

@@ -9,8 +9,7 @@
  * we put the segment information here.
  */
 typedef struct {
-	void *ldt;
+	struct ldt_struct *ldt;
-	int size;
 
 #ifdef CONFIG_X86_64
 	/* True if mm supports a task running in 32 bit compatibility mode. */

arch/x86/include/asm/mmu_context.h

@@ -33,6 +33,50 @@ static inline void load_mm_cr4(struct mm_struct *mm)
 static inline void load_mm_cr4(struct mm_struct *mm) {}
 #endif
 
+/*
+ * ldt_structs can be allocated, used, and freed, but they are never
+ * modified while live.
+ */
+struct ldt_struct {
+	/*
+	 * Xen requires page-aligned LDTs with special permissions.  This is
+	 * needed to prevent us from installing evil descriptors such as
+	 * call gates.  On native, we could merge the ldt_struct and LDT
+	 * allocations, but it's not worth trying to optimize.
+	 */
+	struct desc_struct *entries;
+	int size;
+};
+
+static inline void load_mm_ldt(struct mm_struct *mm)
+{
+	struct ldt_struct *ldt;
+
+	/* lockless_dereference synchronizes with smp_store_release */
+	ldt = lockless_dereference(mm->context.ldt);
+
+	/*
+	 * Any change to mm->context.ldt is followed by an IPI to all
+	 * CPUs with the mm active.  The LDT will not be freed until
+	 * after the IPI is handled by all such CPUs.  This means that,
+	 * if the ldt_struct changes before we return, the values we see
+	 * will be safe, and the new values will be loaded before we run
+	 * any user code.
+	 *
+	 * NB: don't try to convert this to use RCU without extreme care.
+	 * We would still need IRQs off, because we don't want to change
+	 * the local LDT after an IPI loaded a newer value than the one
+	 * that we can see.
+	 */
+
+	if (unlikely(ldt))
+		set_ldt(ldt->entries, ldt->size);
+	else
+		clear_LDT();
+
+	DEBUG_LOCKS_WARN_ON(preemptible());
+}
+
 /*
  * Used for LDT copy/destruction.
  */
@@ -78,12 +122,12 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 		 * was called and then modify_ldt changed
 		 * prev->context.ldt but suppressed an IPI to this CPU.
 		 * In this case, prev->context.ldt != NULL, because we
-		 * never free an LDT while the mm still exists.  That
+		 * never set context.ldt to NULL while the mm still
-		 * means that next->context.ldt != prev->context.ldt,
+		 * exists.  That means that next->context.ldt !=
-		 * because mms never share an LDT.
+		 * prev->context.ldt, because mms never share an LDT.
 		 */
 		if (unlikely(prev->context.ldt != next->context.ldt))
-			load_LDT_nolock(&next->context);
+			load_mm_ldt(next);
 	}
 #ifdef CONFIG_SMP
 	  else {
@@ -106,7 +150,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 			load_cr3(next->pgd);
 			trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL);
 			load_mm_cr4(next);
-			load_LDT_nolock(&next->context);
+			load_mm_ldt(next);
 		}
 	}
 #endif

arch/x86/kernel/apic/io_apic.c

@@ -943,7 +943,7 @@ static bool mp_check_pin_attr(int irq, struct irq_alloc_info *info)
 	 */
 	if (irq < nr_legacy_irqs() && data->count == 1) {
 		if (info->ioapic_trigger != data->trigger)
-			mp_register_handler(irq, data->trigger);
+			mp_register_handler(irq, info->ioapic_trigger);
 		data->entry.trigger = data->trigger = info->ioapic_trigger;
 		data->entry.polarity = data->polarity = info->ioapic_polarity;
 	}

arch/x86/kernel/cpu/common.c

@@ -1410,7 +1410,7 @@ void cpu_init(void)
 	load_sp0(t, &current->thread);
 	set_tss_desc(cpu, t);
 	load_TR_desc();
-	load_LDT(&init_mm.context);
+	load_mm_ldt(&init_mm);
 
 	clear_all_debug_regs();
 	dbg_restore_debug_regs();
@@ -1459,7 +1459,7 @@ void cpu_init(void)
 	load_sp0(t, thread);
 	set_tss_desc(cpu, t);
 	load_TR_desc();
-	load_LDT(&init_mm.context);
+	load_mm_ldt(&init_mm);
 
 	t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
 

arch/x86/kernel/cpu/perf_event.c

@@ -2179,21 +2179,25 @@ static unsigned long get_segment_base(unsigned int segment)
 	int idx = segment >> 3;
 
 	if ((segment & SEGMENT_TI_MASK) == SEGMENT_LDT) {
+		struct ldt_struct *ldt;
+
 		if (idx > LDT_ENTRIES)
 			return 0;
 
-		if (idx > current->active_mm->context.size)
+		/* IRQs are off, so this synchronizes with smp_store_release */
+		ldt = lockless_dereference(current->active_mm->context.ldt);
+		if (!ldt || idx > ldt->size)
 			return 0;
 
-		desc = current->active_mm->context.ldt;
+		desc = &ldt->entries[idx];
 	} else {
 		if (idx > GDT_ENTRIES)
 			return 0;
 
-		desc = raw_cpu_ptr(gdt_page.gdt);
+		desc = raw_cpu_ptr(gdt_page.gdt) + idx;
 	}
 
-	return get_desc_base(desc + idx);
+	return get_desc_base(desc);
 }
 
 #ifdef CONFIG_COMPAT

arch/x86/kernel/ldt.c

@@ -12,6 +12,7 @@
 #include <linux/string.h>
 #include <linux/mm.h>
 #include <linux/smp.h>
+#include <linux/slab.h>
 #include <linux/vmalloc.h>
 #include <linux/uaccess.h>
 
@@ -20,82 +21,82 @@
 #include <asm/mmu_context.h>
 #include <asm/syscalls.h>
 
-#ifdef CONFIG_SMP
+/* context.lock is held for us, so we don't need any locking. */
 static void flush_ldt(void *current_mm)
 {
-	if (current->active_mm == current_mm)
+	mm_context_t *pc;
-		load_LDT(&current->active_mm->context);
+
+	if (current->active_mm != current_mm)
+		return;
+
+	pc = &current->active_mm->context;
+	set_ldt(pc->ldt->entries, pc->ldt->size);
 }
-#endif
 
-static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
+/* The caller must call finalize_ldt_struct on the result. LDT starts zeroed. */
+static struct ldt_struct *alloc_ldt_struct(int size)
 {
-	void *oldldt, *newldt;
+	struct ldt_struct *new_ldt;
-	int oldsize;
+	int alloc_size;
 
-	if (mincount <= pc->size)
+	if (size > LDT_ENTRIES)
-		return 0;
+		return NULL;
-	oldsize = pc->size;
+
-	mincount = (mincount + (PAGE_SIZE / LDT_ENTRY_SIZE - 1)) &
+	new_ldt = kmalloc(sizeof(struct ldt_struct), GFP_KERNEL);
-			(~(PAGE_SIZE / LDT_ENTRY_SIZE - 1));
+	if (!new_ldt)
-	if (mincount * LDT_ENTRY_SIZE > PAGE_SIZE)
+		return NULL;
-		newldt = vmalloc(mincount * LDT_ENTRY_SIZE);
+
+	BUILD_BUG_ON(LDT_ENTRY_SIZE != sizeof(struct desc_struct));
+	alloc_size = size * LDT_ENTRY_SIZE;
+
+	/*
+	 * Xen is very picky: it requires a page-aligned LDT that has no
+	 * trailing nonzero bytes in any page that contains LDT descriptors.
+	 * Keep it simple: zero the whole allocation and never allocate less
+	 * than PAGE_SIZE.
+	 */
+	if (alloc_size > PAGE_SIZE)
+		new_ldt->entries = vzalloc(alloc_size);
 	else
-		newldt = (void *)__get_free_page(GFP_KERNEL);
+		new_ldt->entries = kzalloc(PAGE_SIZE, GFP_KERNEL);
 
-	if (!newldt)
+	if (!new_ldt->entries) {
-		return -ENOMEM;
+		kfree(new_ldt);
-
+		return NULL;
-	if (oldsize)
-		memcpy(newldt, pc->ldt, oldsize * LDT_ENTRY_SIZE);
-	oldldt = pc->ldt;
-	memset(newldt + oldsize * LDT_ENTRY_SIZE, 0,
-	       (mincount - oldsize) * LDT_ENTRY_SIZE);
-
-	paravirt_alloc_ldt(newldt, mincount);
-
-#ifdef CONFIG_X86_64
-	/* CHECKME: Do we really need this ? */
-	wmb();
-#endif
-	pc->ldt = newldt;
-	wmb();
-	pc->size = mincount;
-	wmb();
-
-	if (reload) {
-#ifdef CONFIG_SMP
-		preempt_disable();
-		load_LDT(pc);
-		if (!cpumask_equal(mm_cpumask(current->mm),
-				   cpumask_of(smp_processor_id())))
-			smp_call_function(flush_ldt, current->mm, 1);
-		preempt_enable();
-#else
-		load_LDT(pc);
-#endif
 	}
-	if (oldsize) {
+
-		paravirt_free_ldt(oldldt, oldsize);
+	new_ldt->size = size;
-		if (oldsize * LDT_ENTRY_SIZE > PAGE_SIZE)
+	return new_ldt;
-			vfree(oldldt);
-		else
-			put_page(virt_to_page(oldldt));
-	}
-	return 0;
 }
 
-static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
+/* After calling this, the LDT is immutable. */
+static void finalize_ldt_struct(struct ldt_struct *ldt)
 {
-	int err = alloc_ldt(new, old->size, 0);
+	paravirt_alloc_ldt(ldt->entries, ldt->size);
-	int i;
+}
 
-	if (err < 0)
+/* context.lock is held */
-		return err;
+static void install_ldt(struct mm_struct *current_mm,
+			struct ldt_struct *ldt)
+{
+	/* Synchronizes with lockless_dereference in load_mm_ldt. */
+	smp_store_release(&current_mm->context.ldt, ldt);
 
-	for (i = 0; i < old->size; i++)
+	/* Activate the LDT for all CPUs using current_mm. */
-		write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE);
+	on_each_cpu_mask(mm_cpumask(current_mm), flush_ldt, current_mm, true);
-	return 0;
+}
+
+static void free_ldt_struct(struct ldt_struct *ldt)
+{
+	if (likely(!ldt))
+		return;
+
+	paravirt_free_ldt(ldt->entries, ldt->size);
+	if (ldt->size * LDT_ENTRY_SIZE > PAGE_SIZE)
+		vfree(ldt->entries);
+	else
+		kfree(ldt->entries);
+	kfree(ldt);
 }
 
 /*
@@ -104,17 +105,37 @@ static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
  */
 int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
 {
+	struct ldt_struct *new_ldt;
 	struct mm_struct *old_mm;
 	int retval = 0;
 
 	mutex_init(&mm->context.lock);
-	mm->context.size = 0;
 	old_mm = current->mm;
-	if (old_mm && old_mm->context.size > 0) {
+	if (!old_mm) {
-		mutex_lock(&old_mm->context.lock);
+		mm->context.ldt = NULL;
-		retval = copy_ldt(&mm->context, &old_mm->context);
+		return 0;
-		mutex_unlock(&old_mm->context.lock);
 	}
+
+	mutex_lock(&old_mm->context.lock);
+	if (!old_mm->context.ldt) {
+		mm->context.ldt = NULL;
+		goto out_unlock;
+	}
+
+	new_ldt = alloc_ldt_struct(old_mm->context.ldt->size);
+	if (!new_ldt) {
+		retval = -ENOMEM;
+		goto out_unlock;
+	}
+
+	memcpy(new_ldt->entries, old_mm->context.ldt->entries,
+	       new_ldt->size * LDT_ENTRY_SIZE);
+	finalize_ldt_struct(new_ldt);
+
+	mm->context.ldt = new_ldt;
+
+out_unlock:
+	mutex_unlock(&old_mm->context.lock);
 	return retval;
 }
 
@@ -125,53 +146,47 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
  */
 void destroy_context(struct mm_struct *mm)
 {
-	if (mm->context.size) {
+	free_ldt_struct(mm->context.ldt);
-#ifdef CONFIG_X86_32
+	mm->context.ldt = NULL;
-		/* CHECKME: Can this ever happen ? */
-		if (mm == current->active_mm)
-			clear_LDT();
-#endif
-		paravirt_free_ldt(mm->context.ldt, mm->context.size);
-		if (mm->context.size * LDT_ENTRY_SIZE > PAGE_SIZE)
-			vfree(mm->context.ldt);
-		else
-			put_page(virt_to_page(mm->context.ldt));
-		mm->context.size = 0;
-	}
 }
 
 static int read_ldt(void __user *ptr, unsigned long bytecount)
 {
-	int err;
+	int retval;
 	unsigned long size;
 	struct mm_struct *mm = current->mm;
 
-	if (!mm->context.size)
+	mutex_lock(&mm->context.lock);
-		return 0;
+
+	if (!mm->context.ldt) {
+		retval = 0;
+		goto out_unlock;
+	}
+
 	if (bytecount > LDT_ENTRY_SIZE * LDT_ENTRIES)
 		bytecount = LDT_ENTRY_SIZE * LDT_ENTRIES;
 
-	mutex_lock(&mm->context.lock);
+	size = mm->context.ldt->size * LDT_ENTRY_SIZE;
-	size = mm->context.size * LDT_ENTRY_SIZE;
 	if (size > bytecount)
 		size = bytecount;
 
-	err = 0;
+	if (copy_to_user(ptr, mm->context.ldt->entries, size)) {
-	if (copy_to_user(ptr, mm->context.ldt, size))
+		retval = -EFAULT;
-		err = -EFAULT;
+		goto out_unlock;
-	mutex_unlock(&mm->context.lock);
+	}
-	if (err < 0)
+
-		goto error_return;
 	if (size != bytecount) {
-		/* zero-fill the rest */
+		/* Zero-fill the rest and pretend we read bytecount bytes. */
-		if (clear_user(ptr + size, bytecount - size) != 0) {
+		if (clear_user(ptr + size, bytecount - size)) {
-			err = -EFAULT;
+			retval = -EFAULT;
-			goto error_return;
+			goto out_unlock;
 		}
 	}
-	return bytecount;
+	retval = bytecount;
-error_return:
+
-	return err;
+out_unlock:
+	mutex_unlock(&mm->context.lock);
+	return retval;
 }
 
 static int read_default_ldt(void __user *ptr, unsigned long bytecount)
@@ -195,6 +210,8 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
 	struct desc_struct ldt;
 	int error;
 	struct user_desc ldt_info;
+	int oldsize, newsize;
+	struct ldt_struct *new_ldt, *old_ldt;
 
 	error = -EINVAL;
 	if (bytecount != sizeof(ldt_info))
@@ -213,34 +230,39 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
 			goto out;
 	}
 
-	mutex_lock(&mm->context.lock);
+	if ((oldmode && !ldt_info.base_addr && !ldt_info.limit) ||
-	if (ldt_info.entry_number >= mm->context.size) {
+	    LDT_empty(&ldt_info)) {
-		error = alloc_ldt(&current->mm->context,
+		/* The user wants to clear the entry. */
-				  ldt_info.entry_number + 1, 1);
+		memset(&ldt, 0, sizeof(ldt));
-		if (error < 0)
+	} else {
-			goto out_unlock;
+		if (!IS_ENABLED(CONFIG_X86_16BIT) && !ldt_info.seg_32bit) {
-	}
+			error = -EINVAL;
-
+			goto out;
-	/* Allow LDTs to be cleared by the user. */
-	if (ldt_info.base_addr == 0 && ldt_info.limit == 0) {
-		if (oldmode || LDT_empty(&ldt_info)) {
-			memset(&ldt, 0, sizeof(ldt));
-			goto install;
 		}
+
+		fill_ldt(&ldt, &ldt_info);
+		if (oldmode)
+			ldt.avl = 0;
 	}
 
-	if (!IS_ENABLED(CONFIG_X86_16BIT) && !ldt_info.seg_32bit) {
+	mutex_lock(&mm->context.lock);
-		error = -EINVAL;
+
+	old_ldt = mm->context.ldt;
+	oldsize = old_ldt ? old_ldt->size : 0;
+	newsize = max((int)(ldt_info.entry_number + 1), oldsize);
+
+	error = -ENOMEM;
+	new_ldt = alloc_ldt_struct(newsize);
+	if (!new_ldt)
 		goto out_unlock;
-	}
 
-	fill_ldt(&ldt, &ldt_info);
+	if (old_ldt)
-	if (oldmode)
+		memcpy(new_ldt->entries, old_ldt->entries, oldsize * LDT_ENTRY_SIZE);
-		ldt.avl = 0;
+	new_ldt->entries[ldt_info.entry_number] = ldt;
+	finalize_ldt_struct(new_ldt);
 
-	/* Install the new entry ...  */
+	install_ldt(mm, new_ldt);
-install:
+	free_ldt_struct(old_ldt);
-	write_ldt_entry(mm->context.ldt, ldt_info.entry_number, &ldt);
 	error = 0;
 
 out_unlock:

arch/x86/kernel/process_64.c

@@ -121,11 +121,11 @@ void __show_regs(struct pt_regs *regs, int all)
 void release_thread(struct task_struct *dead_task)
 {
 	if (dead_task->mm) {
-		if (dead_task->mm->context.size) {
+		if (dead_task->mm->context.ldt) {
 			pr_warn("WARNING: dead process %s still has LDT? <%p/%d>\n",
 				dead_task->comm,
 				dead_task->mm->context.ldt,
-				dead_task->mm->context.size);
+				dead_task->mm->context.ldt->size);
 			BUG();
 		}
 	}

arch/x86/kernel/step.c

@@ -5,6 +5,7 @@
 #include <linux/mm.h>
 #include <linux/ptrace.h>
 #include <asm/desc.h>
+#include <asm/mmu_context.h>
 
 unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs)
 {
@@ -30,10 +31,11 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re
 		seg &= ~7UL;
 
 		mutex_lock(&child->mm->context.lock);
-		if (unlikely((seg >> 3) >= child->mm->context.size))
+		if (unlikely(!child->mm->context.ldt ||
+			     (seg >> 3) >= child->mm->context.ldt->size))
 			addr = -1L; /* bogus selector, access would fault */
 		else {
-			desc = child->mm->context.ldt + seg;
+			desc = &child->mm->context.ldt->entries[seg];
 			base = get_desc_base(desc);
 
 			/* 16-bit code segment? */

arch/x86/net/bpf_jit_comp.c

@@ -269,7 +269,7 @@ static void emit_bpf_tail_call(u8 **pprog)
 	EMIT4(0x48, 0x8B, 0x46,                   /* mov rax, qword ptr [rsi + 16] */
 	      offsetof(struct bpf_array, map.max_entries));
 	EMIT3(0x48, 0x39, 0xD0);                  /* cmp rax, rdx */
-#define OFFSET1 44 /* number of bytes to jump */
+#define OFFSET1 47 /* number of bytes to jump */
 	EMIT2(X86_JBE, OFFSET1);                  /* jbe out */
 	label1 = cnt;
 
@@ -278,15 +278,15 @@ static void emit_bpf_tail_call(u8 **pprog)
 	 */
 	EMIT2_off32(0x8B, 0x85, -STACKSIZE + 36); /* mov eax, dword ptr [rbp - 516] */
 	EMIT3(0x83, 0xF8, MAX_TAIL_CALL_CNT);     /* cmp eax, MAX_TAIL_CALL_CNT */
-#define OFFSET2 33
+#define OFFSET2 36
 	EMIT2(X86_JA, OFFSET2);                   /* ja out */
 	label2 = cnt;
 	EMIT3(0x83, 0xC0, 0x01);                  /* add eax, 1 */
 	EMIT2_off32(0x89, 0x85, -STACKSIZE + 36); /* mov dword ptr [rbp - 516], eax */
 
 	/* prog = array->prog[index]; */
-	EMIT4(0x48, 0x8D, 0x44, 0xD6);            /* lea rax, [rsi + rdx * 8 + 0x50] */
+	EMIT4_off32(0x48, 0x8D, 0x84, 0xD6,       /* lea rax, [rsi + rdx * 8 + offsetof(...)] */
-	EMIT1(offsetof(struct bpf_array, prog));
+		    offsetof(struct bpf_array, prog));
 	EMIT3(0x48, 0x8B, 0x00);                  /* mov rax, qword ptr [rax] */
 
 	/* if (prog == NULL)

arch/x86/platform/efi/efi.c

@@ -972,6 +972,11 @@ u64 efi_mem_attributes(unsigned long phys_addr)
 
 static int __init arch_parse_efi_cmdline(char *str)
 {
+	if (!str) {
+		pr_warn("need at least one option\n");
+		return -EINVAL;
+	}
+
 	if (parse_option_str(str, "old_map"))
 		set_bit(EFI_OLD_MEMMAP, &efi.flags);
 	if (parse_option_str(str, "debug"))

arch/x86/power/cpu.c

@@ -22,6 +22,7 @@
 #include <asm/fpu/internal.h>
 #include <asm/debugreg.h>
 #include <asm/cpu.h>
+#include <asm/mmu_context.h>
 
 #ifdef CONFIG_X86_32
 __visible unsigned long saved_context_ebx;
@@ -153,7 +154,7 @@ static void fix_processor_context(void)
 	syscall_init();				/* This sets MSR_*STAR and related */
 #endif
 	load_TR_desc();				/* This does ltr */
-	load_LDT(&current->active_mm->context);	/* This does lldt */
+	load_mm_ldt(current->active_mm);	/* This does lldt */
 
 	fpu__resume_cpu();
 }

arch/x86/xen/enlighten.c

@@ -483,6 +483,7 @@ static void set_aliased_prot(void *v, pgprot_t prot)
 	pte_t pte;
 	unsigned long pfn;
 	struct page *page;
+	unsigned char dummy;
 
 	ptep = lookup_address((unsigned long)v, &level);
 	BUG_ON(ptep == NULL);
@@ -492,6 +493,32 @@ static void set_aliased_prot(void *v, pgprot_t prot)
 
 	pte = pfn_pte(pfn, prot);
 
+	/*
+	 * Careful: update_va_mapping() will fail if the virtual address
+	 * we're poking isn't populated in the page tables.  We don't
+	 * need to worry about the direct map (that's always in the page
+	 * tables), but we need to be careful about vmap space.  In
+	 * particular, the top level page table can lazily propagate
+	 * entries between processes, so if we've switched mms since we
+	 * vmapped the target in the first place, we might not have the
+	 * top-level page table entry populated.
+	 *
+	 * We disable preemption because we want the same mm active when
+	 * we probe the target and when we issue the hypercall.  We'll
+	 * have the same nominal mm, but if we're a kernel thread, lazy
+	 * mm dropping could change our pgd.
+	 *
+	 * Out of an abundance of caution, this uses __get_user() to fault
+	 * in the target address just in case there's some obscure case
+	 * in which the target address isn't readable.
+	 */
+
+	preempt_disable();
+
+	pagefault_disable();	/* Avoid warnings due to being atomic. */
+	__get_user(dummy, (unsigned char __user __force *)v);
+	pagefault_enable();
+
 	if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0))
 		BUG();
 
@@ -503,6 +530,8 @@ static void set_aliased_prot(void *v, pgprot_t prot)
 				BUG();
 	} else
 		kmap_flush_unused();
+
+	preempt_enable();
 }
 
 static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries)
@@ -510,6 +539,17 @@ static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries)
 	const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE;
 	int i;
 
+	/*
+	 * We need to mark the all aliases of the LDT pages RO.  We
+	 * don't need to call vm_flush_aliases(), though, since that's
+	 * only responsible for flushing aliases out the TLBs, not the
+	 * page tables, and Xen will flush the TLB for us if needed.
+	 *
+	 * To avoid confusing future readers: none of this is necessary
+	 * to load the LDT.  The hypervisor only checks this when the
+	 * LDT is faulted in due to subsequent descriptor access.
+	 */
+
 	for(i = 0; i < entries; i += entries_per_page)
 		set_aliased_prot(ldt + i, PAGE_KERNEL_RO);
 }

drivers/acpi/device_pm.c

@@ -231,7 +231,7 @@ int acpi_device_set_power(struct acpi_device *device, int state)
 		dev_warn(&device->dev, "Failed to change power state to %s\n",
 			 acpi_power_state_string(state));
 	} else {
-		device->power.state = state;
+		device->power.state = target_state;
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
 				  "Device [%s] transitioned to %s\n",
 				  device->pnp.bus_id,

drivers/cpufreq/cpufreq-dt.c

@@ -36,6 +36,12 @@ struct private_data {
 	unsigned int voltage_tolerance; /* in percentage */
 };
 
+static struct freq_attr *cpufreq_dt_attr[] = {
+	&cpufreq_freq_attr_scaling_available_freqs,
+	NULL,   /* Extra space for boost-attr if required */
+	NULL,
+};
+
 static int set_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	struct dev_pm_opp *opp;
@@ -184,7 +190,6 @@ try_again:
 
 static int cpufreq_init(struct cpufreq_policy *policy)
 {
-	struct cpufreq_dt_platform_data *pd;
 	struct cpufreq_frequency_table *freq_table;
 	struct device_node *np;
 	struct private_data *priv;
@@ -193,6 +198,7 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 	struct clk *cpu_clk;
 	unsigned long min_uV = ~0, max_uV = 0;
 	unsigned int transition_latency;
+	bool need_update = false;
 	int ret;
 
 	ret = allocate_resources(policy->cpu, &cpu_dev, &cpu_reg, &cpu_clk);
@@ -208,8 +214,47 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 		goto out_put_reg_clk;
 	}
 
-	/* OPPs might be populated at runtime, don't check for error here */
+	/* Get OPP-sharing information from "operating-points-v2" bindings */
-	of_init_opp_table(cpu_dev);
+	ret = of_get_cpus_sharing_opps(cpu_dev, policy->cpus);
+	if (ret) {
+		/*
+		 * operating-points-v2 not supported, fallback to old method of
+		 * finding shared-OPPs for backward compatibility.
+		 */
+		if (ret == -ENOENT)
+			need_update = true;
+		else
+			goto out_node_put;
+	}
+
+	/*
+	 * Initialize OPP tables for all policy->cpus. They will be shared by
+	 * all CPUs which have marked their CPUs shared with OPP bindings.
+	 *
+	 * For platforms not using operating-points-v2 bindings, we do this
+	 * before updating policy->cpus. Otherwise, we will end up creating
+	 * duplicate OPPs for policy->cpus.
+	 *
+	 * OPPs might be populated at runtime, don't check for error here
+	 */
+	of_cpumask_init_opp_table(policy->cpus);
+
+	if (need_update) {
+		struct cpufreq_dt_platform_data *pd = cpufreq_get_driver_data();
+
+		if (!pd || !pd->independent_clocks)
+			cpumask_setall(policy->cpus);
+
+		/*
+		 * OPP tables are initialized only for policy->cpu, do it for
+		 * others as well.
+		 */
+		set_cpus_sharing_opps(cpu_dev, policy->cpus);
+
+		of_property_read_u32(np, "clock-latency", &transition_latency);
+	} else {
+		transition_latency = dev_pm_opp_get_max_clock_latency(cpu_dev);
+	}
 
 	/*
 	 * But we need OPP table to function so if it is not there let's
@@ -230,7 +275,7 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 
 	of_property_read_u32(np, "voltage-tolerance", &priv->voltage_tolerance);
 
-	if (of_property_read_u32(np, "clock-latency", &transition_latency))
+	if (!transition_latency)
 		transition_latency = CPUFREQ_ETERNAL;
 
 	if (!IS_ERR(cpu_reg)) {
@@ -291,11 +336,16 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 		goto out_free_cpufreq_table;
 	}
 
-	policy->cpuinfo.transition_latency = transition_latency;
+	/* Support turbo/boost mode */
+	if (policy_has_boost_freq(policy)) {
+		/* This gets disabled by core on driver unregister */
+		ret = cpufreq_enable_boost_support();
+		if (ret)
+			goto out_free_cpufreq_table;
+		cpufreq_dt_attr[1] = &cpufreq_freq_attr_scaling_boost_freqs;
+	}
 
-	pd = cpufreq_get_driver_data();
+	policy->cpuinfo.transition_latency = transition_latency;
-	if (!pd || !pd->independent_clocks)
-		cpumask_setall(policy->cpus);
 
 	of_node_put(np);
 
@@ -306,7 +356,8 @@ out_free_cpufreq_table:
 out_free_priv:
 	kfree(priv);
 out_free_opp:
-	of_free_opp_table(cpu_dev);
+	of_cpumask_free_opp_table(policy->cpus);
+out_node_put:
 	of_node_put(np);
 out_put_reg_clk:
 	clk_put(cpu_clk);
@@ -322,7 +373,7 @@ static int cpufreq_exit(struct cpufreq_policy *policy)
 
 	cpufreq_cooling_unregister(priv->cdev);
 	dev_pm_opp_free_cpufreq_table(priv->cpu_dev, &policy->freq_table);
-	of_free_opp_table(priv->cpu_dev);
+	of_cpumask_free_opp_table(policy->related_cpus);
 	clk_put(policy->clk);
 	if (!IS_ERR(priv->cpu_reg))
 		regulator_put(priv->cpu_reg);
@@ -367,7 +418,7 @@ static struct cpufreq_driver dt_cpufreq_driver = {
 	.exit = cpufreq_exit,
 	.ready = cpufreq_ready,
 	.name = "cpufreq-dt",
-	.attr = cpufreq_generic_attr,
+	.attr = cpufreq_dt_attr,
 };
 
 static int dt_cpufreq_probe(struct platform_device *pdev)

drivers/cpufreq/cpufreq.c

@@ -2412,6 +2412,49 @@ int cpufreq_boost_supported(void)
 }
 EXPORT_SYMBOL_GPL(cpufreq_boost_supported);
 
+static int create_boost_sysfs_file(void)
+{
+	int ret;
+
+	if (!cpufreq_boost_supported())
+		return 0;
+
+	/*
+	 * Check if driver provides function to enable boost -
+	 * if not, use cpufreq_boost_set_sw as default
+	 */
+	if (!cpufreq_driver->set_boost)
+		cpufreq_driver->set_boost = cpufreq_boost_set_sw;
+
+	ret = cpufreq_sysfs_create_file(&boost.attr);
+	if (ret)
+		pr_err("%s: cannot register global BOOST sysfs file\n",
+		       __func__);
+
+	return ret;
+}
+
+static void remove_boost_sysfs_file(void)
+{
+	if (cpufreq_boost_supported())
+		cpufreq_sysfs_remove_file(&boost.attr);
+}
+
+int cpufreq_enable_boost_support(void)
+{
+	if (!cpufreq_driver)
+		return -EINVAL;
+
+	if (cpufreq_boost_supported())
+		return 0;
+
+	cpufreq_driver->boost_supported = true;
+
+	/* This will get removed on driver unregister */
+	return create_boost_sysfs_file();
+}
+EXPORT_SYMBOL_GPL(cpufreq_enable_boost_support);
+
 int cpufreq_boost_enabled(void)
 {
 	return cpufreq_driver->boost_enabled;
@@ -2465,21 +2508,9 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
 	if (driver_data->setpolicy)
 		driver_data->flags |= CPUFREQ_CONST_LOOPS;
 
-	if (cpufreq_boost_supported()) {
+	ret = create_boost_sysfs_file();
-		/*
+	if (ret)
-		 * Check if driver provides function to enable boost -
+		goto err_null_driver;
-		 * if not, use cpufreq_boost_set_sw as default
-		 */
-		if (!cpufreq_driver->set_boost)
-			cpufreq_driver->set_boost = cpufreq_boost_set_sw;
-
-		ret = cpufreq_sysfs_create_file(&boost.attr);
-		if (ret) {
-			pr_err("%s: cannot register global BOOST sysfs file\n",
-			       __func__);
-			goto err_null_driver;
-		}
-	}
 
 	ret = subsys_interface_register(&cpufreq_interface);
 	if (ret)
@@ -2503,8 +2534,7 @@ out:
 err_if_unreg:
 	subsys_interface_unregister(&cpufreq_interface);
 err_boost_unreg:
-	if (cpufreq_boost_supported())
+	remove_boost_sysfs_file();
-		cpufreq_sysfs_remove_file(&boost.attr);
 err_null_driver:
 	write_lock_irqsave(&cpufreq_driver_lock, flags);
 	cpufreq_driver = NULL;
@@ -2533,9 +2563,7 @@ int cpufreq_unregister_driver(struct cpufreq_driver *driver)
 	/* Protect against concurrent cpu hotplug */
 	get_online_cpus();
 	subsys_interface_unregister(&cpufreq_interface);
-	if (cpufreq_boost_supported())
+	remove_boost_sysfs_file();
-		cpufreq_sysfs_remove_file(&boost.attr);
-
 	unregister_hotcpu_notifier(&cpufreq_cpu_notifier);
 
 	write_lock_irqsave(&cpufreq_driver_lock, flags);

drivers/cpufreq/cpufreq_opp.c

@@ -75,6 +75,10 @@ int dev_pm_opp_init_cpufreq_table(struct device *dev,
 		}
 		freq_table[i].driver_data = i;
 		freq_table[i].frequency = rate / 1000;
+
+		/* Is Boost/turbo opp ? */
+		if (dev_pm_opp_is_turbo(opp))
+			freq_table[i].flags = CPUFREQ_BOOST_FREQ;
 	}
 
 	freq_table[i].driver_data = i;

drivers/cpufreq/freq_table.c

@@ -18,6 +18,21 @@
  *                     FREQUENCY TABLE HELPERS                       *
  *********************************************************************/
 
+bool policy_has_boost_freq(struct cpufreq_policy *policy)
+{
+	struct cpufreq_frequency_table *pos, *table = policy->freq_table;
+
+	if (!table)
+		return false;
+
+	cpufreq_for_each_valid_entry(pos, table)
+		if (pos->flags & CPUFREQ_BOOST_FREQ)
+			return true;
+
+	return false;
+}
+EXPORT_SYMBOL_GPL(policy_has_boost_freq);
+
 int cpufreq_frequency_table_cpuinfo(struct cpufreq_policy *policy,
 				    struct cpufreq_frequency_table *table)
 {

drivers/dma/at_hdmac.c

@@ -48,6 +48,8 @@
 	BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |\
 	BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
 
+#define ATC_MAX_DSCR_TRIALS	10
+
 /*
  * Initial number of descriptors to allocate for each channel. This could
  * be increased during dma usage.
@@ -285,28 +287,19 @@ static struct at_desc *atc_get_desc_by_cookie(struct at_dma_chan *atchan,
  *
  * @current_len: the number of bytes left before reading CTRLA
  * @ctrla: the value of CTRLA
- * @desc: the descriptor containing the transfer width
  */
-static inline int atc_calc_bytes_left(int current_len, u32 ctrla,
+static inline int atc_calc_bytes_left(int current_len, u32 ctrla)
-					struct at_desc *desc)
 {
-	return current_len - ((ctrla & ATC_BTSIZE_MAX) << desc->tx_width);
+	u32 btsize = (ctrla & ATC_BTSIZE_MAX);
-}
+	u32 src_width = ATC_REG_TO_SRC_WIDTH(ctrla);
 
-/**
+	/*
- * atc_calc_bytes_left_from_reg - calculates the number of bytes left according
+	 * According to the datasheet, when reading the Control A Register
- * to the current value of CTRLA.
+	 * (ctrla), the Buffer Transfer Size (btsize) bitfield refers to the
- *
+	 * number of transfers completed on the Source Interface.
- * @current_len: the number of bytes left before reading CTRLA
+	 * So btsize is always a number of source width transfers.
- * @atchan: the channel to read CTRLA for
+	 */
- * @desc: the descriptor containing the transfer width
+	return current_len - (btsize << src_width);
- */
-static inline int atc_calc_bytes_left_from_reg(int current_len,
-			struct at_dma_chan *atchan, struct at_desc *desc)
-{
-	u32 ctrla = channel_readl(atchan, CTRLA);
-
-	return atc_calc_bytes_left(current_len, ctrla, desc);
 }
 
 /**
@@ -320,7 +313,7 @@ static int atc_get_bytes_left(struct dma_chan *chan, dma_cookie_t cookie)
 	struct at_desc *desc_first = atc_first_active(atchan);
 	struct at_desc *desc;
 	int ret;
-	u32 ctrla, dscr;
+	u32 ctrla, dscr, trials;
 
 	/*
 	 * If the cookie doesn't match to the currently running transfer then
@@ -346,15 +339,82 @@ static int atc_get_bytes_left(struct dma_chan *chan, dma_cookie_t cookie)
 		 * the channel's DSCR register and compare it against the value
 		 * of the hardware linked list structure of each child
 		 * descriptor.
+		 *
+		 * The CTRLA register provides us with the amount of data
+		 * already read from the source for the current child
+		 * descriptor. So we can compute a more accurate residue by also
+		 * removing the number of bytes corresponding to this amount of
+		 * data.
+		 *
+		 * However, the DSCR and CTRLA registers cannot be read both
+		 * atomically. Hence a race condition may occur: the first read
+		 * register may refer to one child descriptor whereas the second
+		 * read may refer to a later child descriptor in the list
+		 * because of the DMA transfer progression inbetween the two
+		 * reads.
+		 *
+		 * One solution could have been to pause the DMA transfer, read
+		 * the DSCR and CTRLA then resume the DMA transfer. Nonetheless,
+		 * this approach presents some drawbacks:
+		 * - If the DMA transfer is paused, RX overruns or TX underruns
+		 *   are more likey to occur depending on the system latency.
+		 *   Taking the USART driver as an example, it uses a cyclic DMA
+		 *   transfer to read data from the Receive Holding Register
+		 *   (RHR) to avoid RX overruns since the RHR is not protected
+		 *   by any FIFO on most Atmel SoCs. So pausing the DMA transfer
+		 *   to compute the residue would break the USART driver design.
+		 * - The atc_pause() function masks interrupts but we'd rather
+		 *   avoid to do so for system latency purpose.
+		 *
+		 * Then we'd rather use another solution: the DSCR is read a
+		 * first time, the CTRLA is read in turn, next the DSCR is read
+		 * a second time. If the two consecutive read values of the DSCR
+		 * are the same then we assume both refers to the very same
+		 * child descriptor as well as the CTRLA value read inbetween
+		 * does. For cyclic tranfers, the assumption is that a full loop
+		 * is "not so fast".
+		 * If the two DSCR values are different, we read again the CTRLA
+		 * then the DSCR till two consecutive read values from DSCR are
+		 * equal or till the maxium trials is reach.
+		 * This algorithm is very unlikely not to find a stable value for
+		 * DSCR.
 		 */
 
-		ctrla = channel_readl(atchan, CTRLA);
-		rmb(); /* ensure CTRLA is read before DSCR */
 		dscr = channel_readl(atchan, DSCR);
+		rmb(); /* ensure DSCR is read before CTRLA */
+		ctrla = channel_readl(atchan, CTRLA);
+		for (trials = 0; trials < ATC_MAX_DSCR_TRIALS; ++trials) {
+			u32 new_dscr;
+
+			rmb(); /* ensure DSCR is read after CTRLA */
+			new_dscr = channel_readl(atchan, DSCR);
+
+			/*
+			 * If the DSCR register value has not changed inside the
+			 * DMA controller since the previous read, we assume
+			 * that both the dscr and ctrla values refers to the
+			 * very same descriptor.
+			 */
+			if (likely(new_dscr == dscr))
+				break;
+
+			/*
+			 * DSCR has changed inside the DMA controller, so the
+			 * previouly read value of CTRLA may refer to an already
+			 * processed descriptor hence could be outdated.
+			 * We need to update ctrla to match the current
+			 * descriptor.
+			 */
+			dscr = new_dscr;
+			rmb(); /* ensure DSCR is read before CTRLA */
+			ctrla = channel_readl(atchan, CTRLA);
+		}
+		if (unlikely(trials >= ATC_MAX_DSCR_TRIALS))
+			return -ETIMEDOUT;
 
 		/* for the first descriptor we can be more accurate */
 		if (desc_first->lli.dscr == dscr)
-			return atc_calc_bytes_left(ret, ctrla, desc_first);
+			return atc_calc_bytes_left(ret, ctrla);
 
 		ret -= desc_first->len;
 		list_for_each_entry(desc, &desc_first->tx_list, desc_node) {
@@ -365,16 +425,14 @@ static int atc_get_bytes_left(struct dma_chan *chan, dma_cookie_t cookie)
 		}
 
 		/*
-		 * For the last descriptor in the chain we can calculate
+		 * For the current descriptor in the chain we can calculate
 		 * the remaining bytes using the channel's register.
-		 * Note that the transfer width of the first and last
-		 * descriptor may differ.
 		 */
-		if (!desc->lli.dscr)
+		ret = atc_calc_bytes_left(ret, ctrla);
-			ret = atc_calc_bytes_left_from_reg(ret, atchan, desc);
 	} else {
 		/* single transfer */
-		ret = atc_calc_bytes_left_from_reg(ret, atchan, desc_first);
+		ctrla = channel_readl(atchan, CTRLA);
+		ret = atc_calc_bytes_left(ret, ctrla);
 	}
 
 	return ret;
@@ -726,7 +784,6 @@ atc_prep_dma_interleaved(struct dma_chan *chan,
 
 	desc->txd.cookie = -EBUSY;
 	desc->total_len = desc->len = len;
-	desc->tx_width = dwidth;
 
 	/* set end-of-link to the last link descriptor of list*/
 	set_desc_eol(desc);
@@ -804,10 +861,6 @@ atc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 	first->txd.cookie = -EBUSY;
 	first->total_len = len;
 
-	/* set transfer width for the calculation of the residue */
-	first->tx_width = src_width;
-	prev->tx_width = src_width;
-
 	/* set end-of-link to the last link descriptor of list*/
 	set_desc_eol(desc);
 
@@ -956,10 +1009,6 @@ atc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 	first->txd.cookie = -EBUSY;
 	first->total_len = total_len;
 
-	/* set transfer width for the calculation of the residue */
-	first->tx_width = reg_width;
-	prev->tx_width = reg_width;
-
 	/* first link descriptor of list is responsible of flags */
 	first->txd.flags = flags; /* client is in control of this ack */
 
@@ -1077,12 +1126,6 @@ atc_prep_dma_sg(struct dma_chan *chan,
 		desc->txd.cookie = 0;
 		desc->len = len;
 
-		/*
-		 * Although we only need the transfer width for the first and
-		 * the last descriptor, its easier to set it to all descriptors.
-		 */
-		desc->tx_width = src_width;
-
 		atc_desc_chain(&first, &prev, desc);
 
 		/* update the lengths and addresses for the next loop cycle */
@@ -1256,7 +1299,6 @@ atc_prep_dma_cyclic(struct dma_chan *chan, dma_addr_t buf_addr, size_t buf_len,
 	/* First descriptor of the chain embedds additional information */
 	first->txd.cookie = -EBUSY;
 	first->total_len = buf_len;
-	first->tx_width = reg_width;
 
 	return &first->txd;
 

drivers/dma/at_hdmac_regs.h

@@ -112,6 +112,7 @@
 #define		ATC_SRC_WIDTH_BYTE	(0x0 << 24)
 #define		ATC_SRC_WIDTH_HALFWORD	(0x1 << 24)
 #define		ATC_SRC_WIDTH_WORD	(0x2 << 24)
+#define		ATC_REG_TO_SRC_WIDTH(r)	(((r) >> 24) & 0x3)
 #define	ATC_DST_WIDTH_MASK	(0x3 << 28)	/* Destination Single Transfer Size */
 #define		ATC_DST_WIDTH(x)	((x) << 28)
 #define		ATC_DST_WIDTH_BYTE	(0x0 << 28)
@@ -182,7 +183,6 @@ struct at_lli {
  * @txd: support for the async_tx api
  * @desc_node: node on the channed descriptors list
  * @len: descriptor byte count
- * @tx_width: transfer width
  * @total_len: total transaction byte count
  */
 struct at_desc {
@@ -194,7 +194,6 @@ struct at_desc {
 	struct dma_async_tx_descriptor	txd;
 	struct list_head		desc_node;
 	size_t				len;
-	u32				tx_width;
 	size_t				total_len;
 
 	/* Interleaved data */

drivers/dma/at_xdmac.c

@@ -359,18 +359,19 @@ static void at_xdmac_start_xfer(struct at_xdmac_chan *atchan,
 	 * descriptor view 2 since some fields of the configuration register
 	 * depend on transfer size and src/dest addresses.
 	 */
-	if (at_xdmac_chan_is_cyclic(atchan)) {
+	if (at_xdmac_chan_is_cyclic(atchan))
 		reg = AT_XDMAC_CNDC_NDVIEW_NDV1;
-		at_xdmac_chan_write(atchan, AT_XDMAC_CC, first->lld.mbr_cfg);
+	else if (first->lld.mbr_ubc & AT_XDMAC_MBR_UBC_NDV3)
-	} else if (first->lld.mbr_ubc & AT_XDMAC_MBR_UBC_NDV3) {
 		reg = AT_XDMAC_CNDC_NDVIEW_NDV3;
-	} else {
+	else
-		/*
-		 * No need to write AT_XDMAC_CC reg, it will be done when the
-		 * descriptor is fecthed.
-		 */
 		reg = AT_XDMAC_CNDC_NDVIEW_NDV2;
-	}
+	/*
+	 * Even if the register will be updated from the configuration in the
+	 * descriptor when using view 2 or higher, the PROT bit won't be set
+	 * properly. This bit can be modified only by using the channel
+	 * configuration register.
+	 */
+	at_xdmac_chan_write(atchan, AT_XDMAC_CC, first->lld.mbr_cfg);
 
 	reg |= AT_XDMAC_CNDC_NDDUP
 	       | AT_XDMAC_CNDC_NDSUP
@@ -681,15 +682,16 @@ at_xdmac_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 			desc->lld.mbr_sa = mem;
 			desc->lld.mbr_da = atchan->sconfig.dst_addr;
 		}
-		desc->lld.mbr_cfg = atchan->cfg;
+		dwidth = at_xdmac_get_dwidth(atchan->cfg);
-		dwidth = at_xdmac_get_dwidth(desc->lld.mbr_cfg);
 		fixed_dwidth = IS_ALIGNED(len, 1 << dwidth)
-			       ? at_xdmac_get_dwidth(desc->lld.mbr_cfg)
+			       ? dwidth
 			       : AT_XDMAC_CC_DWIDTH_BYTE;
 		desc->lld.mbr_ubc = AT_XDMAC_MBR_UBC_NDV2			/* next descriptor view */
 			| AT_XDMAC_MBR_UBC_NDEN					/* next descriptor dst parameter update */
 			| AT_XDMAC_MBR_UBC_NSEN					/* next descriptor src parameter update */
 			| (len >> fixed_dwidth);				/* microblock length */
+		desc->lld.mbr_cfg = (atchan->cfg & ~AT_XDMAC_CC_DWIDTH_MASK) |
+				    AT_XDMAC_CC_DWIDTH(fixed_dwidth);
 		dev_dbg(chan2dev(chan),
 			 "%s: lld: mbr_sa=%pad, mbr_da=%pad, mbr_ubc=0x%08x\n",
 			 __func__, &desc->lld.mbr_sa, &desc->lld.mbr_da, desc->lld.mbr_ubc);

drivers/dma/mv_xor.c

@@ -162,10 +162,11 @@ static void mv_chan_set_mode(struct mv_xor_chan *chan,
 	config &= ~0x7;
 	config |= op_mode;
 
-	if (IS_ENABLED(__BIG_ENDIAN))
+#if defined(__BIG_ENDIAN)
-		config |= XOR_DESCRIPTOR_SWAP;
+	config |= XOR_DESCRIPTOR_SWAP;
-	else
+#else
-		config &= ~XOR_DESCRIPTOR_SWAP;
+	config &= ~XOR_DESCRIPTOR_SWAP;
+#endif
 
 	writel_relaxed(config, XOR_CONFIG(chan));
 	chan->current_type = type;

drivers/dma/pl330.c

@@ -2328,7 +2328,7 @@ static dma_cookie_t pl330_tx_submit(struct dma_async_tx_descriptor *tx)
 			desc->txd.callback = last->txd.callback;
 			desc->txd.callback_param = last->txd.callback_param;
 		}
-		last->last = false;
+		desc->last = false;
 
 		dma_cookie_assign(&desc->txd);
 
@@ -2623,6 +2623,7 @@ pl330_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dst,
 		desc->rqcfg.brst_len = 1;
 
 	desc->rqcfg.brst_len = get_burst_len(desc, len);
+	desc->bytes_requested = len;
 
 	desc->txd.flags = flags;
 

drivers/dma/virt-dma.c

@@ -29,7 +29,7 @@ dma_cookie_t vchan_tx_submit(struct dma_async_tx_descriptor *tx)
 	spin_lock_irqsave(&vc->lock, flags);
 	cookie = dma_cookie_assign(tx);
 
-	list_move_tail(&vd->node, &vc->desc_submitted);
+	list_add_tail(&vd->node, &vc->desc_submitted);
 	spin_unlock_irqrestore(&vc->lock, flags);
 
 	dev_dbg(vc->chan.device->dev, "vchan %p: txd %p[%x]: submitted\n",
@@ -83,10 +83,8 @@ static void vchan_complete(unsigned long arg)
 		cb_data = vd->tx.callback_param;
 
 		list_del(&vd->node);
-		if (async_tx_test_ack(&vd->tx))
+
-			list_add(&vd->node, &vc->desc_allocated);
+		vc->desc_free(vd);
-		else
-			vc->desc_free(vd);
 
 		if (cb)
 			cb(cb_data);
@@ -98,13 +96,9 @@ void vchan_dma_desc_free_list(struct virt_dma_chan *vc, struct list_head *head)
 	while (!list_empty(head)) {
 		struct virt_dma_desc *vd = list_first_entry(head,
 			struct virt_dma_desc, node);
-		if (async_tx_test_ack(&vd->tx)) {
+		list_del(&vd->node);
-			list_move_tail(&vd->node, &vc->desc_allocated);
+		dev_dbg(vc->chan.device->dev, "txd %p: freeing\n", vd);
-		} else {
+		vc->desc_free(vd);
-			dev_dbg(vc->chan.device->dev, "txd %p: freeing\n", vd);
-			list_del(&vd->node);
-			vc->desc_free(vd);
-		}
 	}
 }
 EXPORT_SYMBOL_GPL(vchan_dma_desc_free_list);
@@ -114,7 +108,6 @@ void vchan_init(struct virt_dma_chan *vc, struct dma_device *dmadev)
 	dma_cookie_init(&vc->chan);
 
 	spin_lock_init(&vc->lock);
-	INIT_LIST_HEAD(&vc->desc_allocated);
 	INIT_LIST_HEAD(&vc->desc_submitted);
 	INIT_LIST_HEAD(&vc->desc_issued);
 	INIT_LIST_HEAD(&vc->desc_completed);

drivers/dma/virt-dma.h

@@ -29,7 +29,6 @@ struct virt_dma_chan {
 	spinlock_t lock;
 
 	/* protected by vc.lock */
-	struct list_head desc_allocated;
 	struct list_head desc_submitted;
 	struct list_head desc_issued;
 	struct list_head desc_completed;
@@ -56,16 +55,11 @@ static inline struct dma_async_tx_descriptor *vchan_tx_prep(struct virt_dma_chan
 	struct virt_dma_desc *vd, unsigned long tx_flags)
 {
 	extern dma_cookie_t vchan_tx_submit(struct dma_async_tx_descriptor *);
-	unsigned long flags;
 
 	dma_async_tx_descriptor_init(&vd->tx, &vc->chan);
 	vd->tx.flags = tx_flags;
 	vd->tx.tx_submit = vchan_tx_submit;
 
-	spin_lock_irqsave(&vc->lock, flags);
-	list_add_tail(&vd->node, &vc->desc_allocated);
-	spin_unlock_irqrestore(&vc->lock, flags);
-
 	return &vd->tx;
 }
 
@@ -128,8 +122,7 @@ static inline struct virt_dma_desc *vchan_next_desc(struct virt_dma_chan *vc)
 }
 
 /**
- * vchan_get_all_descriptors - obtain all allocated, submitted and issued
+ * vchan_get_all_descriptors - obtain all submitted and issued descriptors
- *                             descriptors
  * vc: virtual channel to get descriptors from
  * head: list of descriptors found
  *
@@ -141,7 +134,6 @@ static inline struct virt_dma_desc *vchan_next_desc(struct virt_dma_chan *vc)
 static inline void vchan_get_all_descriptors(struct virt_dma_chan *vc,
 	struct list_head *head)
 {
-	list_splice_tail_init(&vc->desc_allocated, head);
 	list_splice_tail_init(&vc->desc_submitted, head);
 	list_splice_tail_init(&vc->desc_issued, head);
 	list_splice_tail_init(&vc->desc_completed, head);
@@ -149,14 +141,11 @@ static inline void vchan_get_all_descriptors(struct virt_dma_chan *vc,
 
 static inline void vchan_free_chan_resources(struct virt_dma_chan *vc)
 {
-	struct virt_dma_desc *vd;
 	unsigned long flags;
 	LIST_HEAD(head);
 
 	spin_lock_irqsave(&vc->lock, flags);
 	vchan_get_all_descriptors(vc, &head);
-	list_for_each_entry(vd, &head, node)
-		async_tx_clear_ack(&vd->tx);
 	spin_unlock_irqrestore(&vc->lock, flags);
 
 	vchan_dma_desc_free_list(vc, &head);

drivers/dma/xgene-dma.c

@@ -111,6 +111,7 @@
 #define XGENE_DMA_MEM_RAM_SHUTDOWN		0xD070
 #define XGENE_DMA_BLK_MEM_RDY			0xD074
 #define XGENE_DMA_BLK_MEM_RDY_VAL		0xFFFFFFFF
+#define XGENE_DMA_RING_CMD_SM_OFFSET		0x8000
 
 /* X-Gene SoC EFUSE csr register and bit defination */
 #define XGENE_SOC_JTAG1_SHADOW			0x18
@@ -1887,6 +1888,8 @@ static int xgene_dma_get_resources(struct platform_device *pdev,
 		return -ENOMEM;
 	}
 
+	pdma->csr_ring_cmd += XGENE_DMA_RING_CMD_SM_OFFSET;
+
 	/* Get efuse csr region */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 3);
 	if (!res) {

drivers/firmware/efi/efi.c

@@ -58,6 +58,11 @@ bool efi_runtime_disabled(void)
 
 static int __init parse_efi_cmdline(char *str)
 {
+	if (!str) {
+		pr_warn("need at least one option\n");
+		return -EINVAL;
+	}
+
 	if (parse_option_str(str, "noruntime"))
 		disable_runtime = true;
 

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -1866,6 +1866,12 @@ typedef void (*amdgpu_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t);
 typedef uint32_t (*amdgpu_block_rreg_t)(struct amdgpu_device*, uint32_t, uint32_t);
 typedef void (*amdgpu_block_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t, uint32_t);
 
+struct amdgpu_ip_block_status {
+	bool valid;
+	bool sw;
+	bool hw;
+};
+
 struct amdgpu_device {
 	struct device			*dev;
 	struct drm_device		*ddev;
@@ -2008,7 +2014,7 @@ struct amdgpu_device {
 
 	const struct amdgpu_ip_block_version *ip_blocks;
 	int				num_ip_blocks;
-	bool				*ip_block_enabled;
+	struct amdgpu_ip_block_status	*ip_block_status;
 	struct mutex	mn_lock;
 	DECLARE_HASHTABLE(mn_hash, 7);
 

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

@@ -1191,8 +1191,9 @@ static int amdgpu_early_init(struct amdgpu_device *adev)
 		return -EINVAL;
 	}
 
-	adev->ip_block_enabled = kcalloc(adev->num_ip_blocks, sizeof(bool), GFP_KERNEL);
+	adev->ip_block_status = kcalloc(adev->num_ip_blocks,
-	if (adev->ip_block_enabled == NULL)
+					sizeof(struct amdgpu_ip_block_status), GFP_KERNEL);
+	if (adev->ip_block_status == NULL)
 		return -ENOMEM;
 
 	if (adev->ip_blocks == NULL) {
@@ -1203,18 +1204,18 @@ static int amdgpu_early_init(struct amdgpu_device *adev)
 	for (i = 0; i < adev->num_ip_blocks; i++) {
 		if ((amdgpu_ip_block_mask & (1 << i)) == 0) {
 			DRM_ERROR("disabled ip block: %d\n", i);
-			adev->ip_block_enabled[i] = false;
+			adev->ip_block_status[i].valid = false;
 		} else {
 			if (adev->ip_blocks[i].funcs->early_init) {
 				r = adev->ip_blocks[i].funcs->early_init((void *)adev);
 				if (r == -ENOENT)
-					adev->ip_block_enabled[i] = false;
+					adev->ip_block_status[i].valid = false;
 				else if (r)
 					return r;
 				else
-					adev->ip_block_enabled[i] = true;
+					adev->ip_block_status[i].valid = true;
 			} else {
-				adev->ip_block_enabled[i] = true;
+				adev->ip_block_status[i].valid = true;
 			}
 		}
 	}
@@ -1227,11 +1228,12 @@ static int amdgpu_init(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		r = adev->ip_blocks[i].funcs->sw_init((void *)adev);
 		if (r)
 			return r;
+		adev->ip_block_status[i].sw = true;
 		/* need to do gmc hw init early so we can allocate gpu mem */
 		if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) {
 			r = amdgpu_vram_scratch_init(adev);
@@ -1243,11 +1245,12 @@ static int amdgpu_init(struct amdgpu_device *adev)
 			r = amdgpu_wb_init(adev);
 			if (r)
 				return r;
+			adev->ip_block_status[i].hw = true;
 		}
 	}
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].sw)
 			continue;
 		/* gmc hw init is done early */
 		if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC)
@@ -1255,6 +1258,7 @@ static int amdgpu_init(struct amdgpu_device *adev)
 		r = adev->ip_blocks[i].funcs->hw_init((void *)adev);
 		if (r)
 			return r;
+		adev->ip_block_status[i].hw = true;
 	}
 
 	return 0;
@@ -1265,7 +1269,7 @@ static int amdgpu_late_init(struct amdgpu_device *adev)
 	int i = 0, r;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		/* enable clockgating to save power */
 		r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev,
@@ -1287,7 +1291,7 @@ static int amdgpu_fini(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].hw)
 			continue;
 		if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) {
 			amdgpu_wb_fini(adev);
@@ -1300,14 +1304,16 @@ static int amdgpu_fini(struct amdgpu_device *adev)
 			return r;
 		r = adev->ip_blocks[i].funcs->hw_fini((void *)adev);
 		/* XXX handle errors */
+		adev->ip_block_status[i].hw = false;
 	}
 
 	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].sw)
 			continue;
 		r = adev->ip_blocks[i].funcs->sw_fini((void *)adev);
 		/* XXX handle errors */
-		adev->ip_block_enabled[i] = false;
+		adev->ip_block_status[i].sw = false;
+		adev->ip_block_status[i].valid = false;
 	}
 
 	return 0;
@@ -1318,7 +1324,7 @@ static int amdgpu_suspend(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		/* ungate blocks so that suspend can properly shut them down */
 		r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev,
@@ -1336,7 +1342,7 @@ static int amdgpu_resume(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		r = adev->ip_blocks[i].funcs->resume(adev);
 		if (r)
@@ -1582,8 +1588,8 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 	amdgpu_fence_driver_fini(adev);
 	amdgpu_fbdev_fini(adev);
 	r = amdgpu_fini(adev);
-	kfree(adev->ip_block_enabled);
+	kfree(adev->ip_block_status);
-	adev->ip_block_enabled = NULL;
+	adev->ip_block_status = NULL;
 	adev->accel_working = false;
 	/* free i2c buses */
 	amdgpu_i2c_fini(adev);

drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c

@@ -449,7 +449,7 @@ out:
  * vital here, so they are not reported back to userspace.
  */
 static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
-				    struct amdgpu_bo_va *bo_va)
+				    struct amdgpu_bo_va *bo_va, uint32_t operation)
 {
 	struct ttm_validate_buffer tv, *entry;
 	struct amdgpu_bo_list_entry *vm_bos;
@@ -485,7 +485,9 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
 	if (r)
 		goto error_unlock;
 
-	r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem);
+
+	if (operation == AMDGPU_VA_OP_MAP)
+		r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem);
 
 error_unlock:
 	mutex_unlock(&bo_va->vm->mutex);
@@ -580,7 +582,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	}
 
 	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE))
-		amdgpu_gem_va_update_vm(adev, bo_va);
+		amdgpu_gem_va_update_vm(adev, bo_va, args->operation);
 
 	drm_gem_object_unreference_unlocked(gobj);
 	return r;

drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c

@@ -180,17 +180,17 @@ int amdgpu_ib_schedule(struct amdgpu_device *adev, unsigned num_ibs,
 	if (vm) {
 		/* do context switch */
 		amdgpu_vm_flush(ring, vm, ib->sync.last_vm_update);
+
+		if (ring->funcs->emit_gds_switch)
+			amdgpu_ring_emit_gds_switch(ring, ib->vm->ids[ring->idx].id,
+						    ib->gds_base, ib->gds_size,
+						    ib->gws_base, ib->gws_size,
+						    ib->oa_base, ib->oa_size);
+
+		if (ring->funcs->emit_hdp_flush)
+			amdgpu_ring_emit_hdp_flush(ring);
 	}
 
-	if (vm && ring->funcs->emit_gds_switch)
-		amdgpu_ring_emit_gds_switch(ring, ib->vm->ids[ring->idx].id,
-					    ib->gds_base, ib->gds_size,
-					    ib->gws_base, ib->gws_size,
-					    ib->oa_base, ib->oa_size);
-
-	if (ring->funcs->emit_hdp_flush)
-		amdgpu_ring_emit_hdp_flush(ring);
-
 	old_ctx = ring->current_ctx;
 	for (i = 0; i < num_ibs; ++i) {
 		ib = &ibs[i];

drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c

@@ -235,7 +235,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 
 		for (i = 0; i < adev->num_ip_blocks; i++) {
 			if (adev->ip_blocks[i].type == type &&
-			    adev->ip_block_enabled[i]) {
+			    adev->ip_block_status[i].valid) {
 				ip.hw_ip_version_major = adev->ip_blocks[i].major;
 				ip.hw_ip_version_minor = adev->ip_blocks[i].minor;
 				ip.capabilities_flags = 0;
@@ -274,7 +274,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 
 		for (i = 0; i < adev->num_ip_blocks; i++)
 			if (adev->ip_blocks[i].type == type &&
-			    adev->ip_block_enabled[i] &&
+			    adev->ip_block_status[i].valid &&
 			    count < AMDGPU_HW_IP_INSTANCE_MAX_COUNT)
 				count++;
 
@@ -416,7 +416,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		return n ? -EFAULT : 0;
 	}
 	case AMDGPU_INFO_DEV_INFO: {
-		struct drm_amdgpu_info_device dev_info;
+		struct drm_amdgpu_info_device dev_info = {};
 		struct amdgpu_cu_info cu_info;
 
 		dev_info.device_id = dev->pdev->device;

drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c

@@ -2561,7 +2561,7 @@ static bool gfx_v7_0_ring_emit_semaphore(struct amdgpu_ring *ring,
  * sheduling on the ring.  This function schedules the IB
  * on the gfx ring for execution by the GPU.
  */
-static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
+static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 				  struct amdgpu_ib *ib)
 {
 	bool need_ctx_switch = ring->current_ctx != ib->ctx;
@@ -2569,15 +2569,10 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 	u32 next_rptr = ring->wptr + 5;
 
 	/* drop the CE preamble IB for the same context */
-	if ((ring->type == AMDGPU_RING_TYPE_GFX) &&
+	if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch)
-	    (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) &&
-	    !need_ctx_switch)
 		return;
 
-	if (ring->type == AMDGPU_RING_TYPE_COMPUTE)
+	if (need_ctx_switch)
-		control |= INDIRECT_BUFFER_VALID;
-
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX)
 		next_rptr += 2;
 
 	next_rptr += 4;
@@ -2588,7 +2583,7 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, next_rptr);
 
 	/* insert SWITCH_BUFFER packet before first IB in the ring frame */
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) {
+	if (need_ctx_switch) {
 		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
 		amdgpu_ring_write(ring, 0);
 	}
@@ -2611,6 +2606,35 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, control);
 }
 
+static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
+				  struct amdgpu_ib *ib)
+{
+	u32 header, control = 0;
+	u32 next_rptr = ring->wptr + 5;
+
+	control |= INDIRECT_BUFFER_VALID;
+	next_rptr += 4;
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM);
+	amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+	amdgpu_ring_write(ring, next_rptr);
+
+	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
+
+	control |= ib->length_dw |
+			   (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+
+	amdgpu_ring_write(ring, header);
+	amdgpu_ring_write(ring,
+#ifdef __BIG_ENDIAN
+					  (2 << 0) |
+#endif
+					  (ib->gpu_addr & 0xFFFFFFFC));
+	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF);
+	amdgpu_ring_write(ring, control);
+}
+
 /**
  * gfx_v7_0_ring_test_ib - basic ring IB test
  *
@@ -5555,7 +5579,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = {
 	.get_wptr = gfx_v7_0_ring_get_wptr_gfx,
 	.set_wptr = gfx_v7_0_ring_set_wptr_gfx,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v7_0_ring_emit_ib,
+	.emit_ib = gfx_v7_0_ring_emit_ib_gfx,
 	.emit_fence = gfx_v7_0_ring_emit_fence_gfx,
 	.emit_semaphore = gfx_v7_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v7_0_ring_emit_vm_flush,
@@ -5571,7 +5595,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = {
 	.get_wptr = gfx_v7_0_ring_get_wptr_compute,
 	.set_wptr = gfx_v7_0_ring_set_wptr_compute,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v7_0_ring_emit_ib,
+	.emit_ib = gfx_v7_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v7_0_ring_emit_fence_compute,
 	.emit_semaphore = gfx_v7_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v7_0_ring_emit_vm_flush,

drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c

@@ -3753,7 +3753,7 @@ static void gfx_v8_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 	amdgpu_ring_write(ring, 0x20); /* poll interval */
 }
 
-static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring,
+static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 				  struct amdgpu_ib *ib)
 {
 	bool need_ctx_switch = ring->current_ctx != ib->ctx;
@@ -3761,15 +3761,10 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring,
 	u32 next_rptr = ring->wptr + 5;
 
 	/* drop the CE preamble IB for the same context */
-	if ((ring->type == AMDGPU_RING_TYPE_GFX) &&
+	if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch)
-	    (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) &&
-	    !need_ctx_switch)
 		return;
 
-	if (ring->type == AMDGPU_RING_TYPE_COMPUTE)
+	if (need_ctx_switch)
-		control |= INDIRECT_BUFFER_VALID;
-
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX)
 		next_rptr += 2;
 
 	next_rptr += 4;
@@ -3780,7 +3775,7 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, next_rptr);
 
 	/* insert SWITCH_BUFFER packet before first IB in the ring frame */
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) {
+	if (need_ctx_switch) {
 		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
 		amdgpu_ring_write(ring, 0);
 	}
@@ -3803,6 +3798,36 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, control);
 }
 
+static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
+				  struct amdgpu_ib *ib)
+{
+	u32 header, control = 0;
+	u32 next_rptr = ring->wptr + 5;
+
+	control |= INDIRECT_BUFFER_VALID;
+
+	next_rptr += 4;
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM);
+	amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+	amdgpu_ring_write(ring, next_rptr);
+
+	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
+
+	control |= ib->length_dw |
+			   (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+
+	amdgpu_ring_write(ring, header);
+	amdgpu_ring_write(ring,
+#ifdef __BIG_ENDIAN
+					  (2 << 0) |
+#endif
+					  (ib->gpu_addr & 0xFFFFFFFC));
+	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF);
+	amdgpu_ring_write(ring, control);
+}
+
 static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr,
 					 u64 seq, unsigned flags)
 {
@@ -4224,7 +4249,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
 	.get_wptr = gfx_v8_0_ring_get_wptr_gfx,
 	.set_wptr = gfx_v8_0_ring_set_wptr_gfx,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v8_0_ring_emit_ib,
+	.emit_ib = gfx_v8_0_ring_emit_ib_gfx,
 	.emit_fence = gfx_v8_0_ring_emit_fence_gfx,
 	.emit_semaphore = gfx_v8_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,
@@ -4240,7 +4265,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
 	.get_wptr = gfx_v8_0_ring_get_wptr_compute,
 	.set_wptr = gfx_v8_0_ring_set_wptr_compute,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v8_0_ring_emit_ib,
+	.emit_ib = gfx_v8_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v8_0_ring_emit_fence_compute,
 	.emit_semaphore = gfx_v8_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,

drivers/gpu/drm/drm_atomic_helper.c

@@ -230,10 +230,12 @@ update_connector_routing(struct drm_atomic_state *state, int conn_idx)
 	}
 
 	connector_state->best_encoder = new_encoder;
-	idx = drm_crtc_index(connector_state->crtc);
+	if (connector_state->crtc) {
+		idx = drm_crtc_index(connector_state->crtc);
 
-	crtc_state = state->crtc_states[idx];
+		crtc_state = state->crtc_states[idx];
-	crtc_state->mode_changed = true;
+		crtc_state->mode_changed = true;
+	}
 
 	DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] using [ENCODER:%d:%s] on [CRTC:%d]\n",
 			 connector->base.id,

drivers/gpu/drm/i915/i915_drv.h

@@ -3303,15 +3303,14 @@ int intel_freq_opcode(struct drm_i915_private *dev_priv, int val);
 #define I915_READ64(reg)	dev_priv->uncore.funcs.mmio_readq(dev_priv, (reg), true)
 
 #define I915_READ64_2x32(lower_reg, upper_reg) ({			\
-		u32 upper = I915_READ(upper_reg);			\
+	u32 upper, lower, tmp;						\
-		u32 lower = I915_READ(lower_reg);			\
+	tmp = I915_READ(upper_reg);					\
-		u32 tmp = I915_READ(upper_reg);				\
+	do {								\
-		if (upper != tmp) {					\
+		upper = tmp;						\
-			upper = tmp;					\
+		lower = I915_READ(lower_reg);				\
-			lower = I915_READ(lower_reg);			\
+		tmp = I915_READ(upper_reg);				\
-			WARN_ON(I915_READ(upper_reg) != upper);		\
+	} while (upper != tmp);						\
-		}							\
+	(u64)upper << 32 | lower; })
-		(u64)upper << 32 | lower; })
 
 #define POSTING_READ(reg)	(void)I915_READ_NOTRACE(reg)
 #define POSTING_READ16(reg)	(void)I915_READ16_NOTRACE(reg)

drivers/gpu/drm/i915/i915_gem_gtt.c

@@ -1923,6 +1923,17 @@ static int ggtt_bind_vma(struct i915_vma *vma,
 		vma->vm->insert_entries(vma->vm, pages,
 					vma->node.start,
 					cache_level, pte_flags);
+
+		/* Note the inconsistency here is due to absence of the
+		 * aliasing ppgtt on gen4 and earlier. Though we always
+		 * request PIN_USER for execbuffer (translated to LOCAL_BIND),
+		 * without the appgtt, we cannot honour that request and so
+		 * must substitute it with a global binding. Since we do this
+		 * behind the upper layers back, we need to explicitly set
+		 * the bound flag ourselves.
+		 */
+		vma->bound |= GLOBAL_BIND;
+
 	}
 
 	if (dev_priv->mm.aliasing_ppgtt && flags & LOCAL_BIND) {

drivers/gpu/drm/i915/i915_gem_tiling.c

@@ -464,7 +464,10 @@ i915_gem_get_tiling(struct drm_device *dev, void *data,
 	}
 
 	/* Hide bit 17 from the user -- see comment in i915_gem_set_tiling */
-	args->phys_swizzle_mode = args->swizzle_mode;
+	if (dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES)
+		args->phys_swizzle_mode = I915_BIT_6_SWIZZLE_UNKNOWN;
+	else
+		args->phys_swizzle_mode = args->swizzle_mode;
 	if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_17)
 		args->swizzle_mode = I915_BIT_6_SWIZZLE_9;
 	if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17)

drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c

@@ -220,13 +220,15 @@ static int mdp4_plane_mode_set(struct drm_plane *plane,
 	uint32_t op_mode = 0;
 	uint32_t phasex_step = MDP4_VG_PHASE_STEP_DEFAULT;
 	uint32_t phasey_step = MDP4_VG_PHASE_STEP_DEFAULT;
-	enum mdp4_frame_format frame_type = mdp4_get_frame_format(fb);
+	enum mdp4_frame_format frame_type;
 
 	if (!(crtc && fb)) {
 		DBG("%s: disabled!", mdp4_plane->name);
 		return 0;
 	}
 
+	frame_type = mdp4_get_frame_format(fb);
+
 	/* src values are in Q16 fixed point, convert to integer: */
 	src_x = src_x >> 16;
 	src_y = src_y >> 16;

drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c

@@ -76,7 +76,20 @@ static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *st
 
 static void mdp5_complete_commit(struct msm_kms *kms, struct drm_atomic_state *state)
 {
+	int i;
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
+	int nplanes = mdp5_kms->dev->mode_config.num_total_plane;
+
+	for (i = 0; i < nplanes; i++) {
+		struct drm_plane *plane = state->planes[i];
+		struct drm_plane_state *plane_state = state->plane_states[i];
+
+		if (!plane)
+			continue;
+
+		mdp5_plane_complete_commit(plane, plane_state);
+	}
+
 	mdp5_disable(mdp5_kms);
 }
 

drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h

@@ -227,6 +227,8 @@ void mdp5_plane_install_properties(struct drm_plane *plane,
 		struct drm_mode_object *obj);
 uint32_t mdp5_plane_get_flush(struct drm_plane *plane);
 void mdp5_plane_complete_flip(struct drm_plane *plane);
+void mdp5_plane_complete_commit(struct drm_plane *plane,
+	struct drm_plane_state *state);
 enum mdp5_pipe mdp5_plane_pipe(struct drm_plane *plane);
 struct drm_plane *mdp5_plane_init(struct drm_device *dev,
 		enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset);

drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c

@@ -31,8 +31,6 @@ struct mdp5_plane {
 
 	uint32_t nformats;
 	uint32_t formats[32];
-
-	bool enabled;
 };
 #define to_mdp5_plane(x) container_of(x, struct mdp5_plane, base)
 
@@ -56,22 +54,6 @@ static bool plane_enabled(struct drm_plane_state *state)
 	return state->fb && state->crtc;
 }
 
-static int mdp5_plane_disable(struct drm_plane *plane)
-{
-	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
-	struct mdp5_kms *mdp5_kms = get_kms(plane);
-	enum mdp5_pipe pipe = mdp5_plane->pipe;
-
-	DBG("%s: disable", mdp5_plane->name);
-
-	if (mdp5_kms) {
-		/* Release the memory we requested earlier from the SMP: */
-		mdp5_smp_release(mdp5_kms->smp, pipe);
-	}
-
-	return 0;
-}
-
 static void mdp5_plane_destroy(struct drm_plane *plane)
 {
 	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
@@ -224,7 +206,6 @@ static void mdp5_plane_atomic_update(struct drm_plane *plane,
 
 	if (!plane_enabled(state)) {
 		to_mdp5_plane_state(state)->pending = true;
-		mdp5_plane_disable(plane);
 	} else if (to_mdp5_plane_state(state)->mode_changed) {
 		int ret;
 		to_mdp5_plane_state(state)->pending = true;
@@ -602,6 +583,20 @@ uint32_t mdp5_plane_get_flush(struct drm_plane *plane)
 	return mdp5_plane->flush_mask;
 }
 
+/* called after vsync in thread context */
+void mdp5_plane_complete_commit(struct drm_plane *plane,
+	struct drm_plane_state *state)
+{
+	struct mdp5_kms *mdp5_kms = get_kms(plane);
+	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
+	enum mdp5_pipe pipe = mdp5_plane->pipe;
+
+	if (!plane_enabled(plane->state)) {
+		DBG("%s: free SMP", mdp5_plane->name);
+		mdp5_smp_release(mdp5_kms->smp, pipe);
+	}
+}
+
 /* initialize plane */
 struct drm_plane *mdp5_plane_init(struct drm_device *dev,
 		enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset)

drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c

@@ -34,22 +34,44 @@
  * and CANNOT be re-allocated (eg: MMB0 and MMB1 both tied to RGB0).
  *
  * For each block that can be dynamically allocated, it can be either
- * free, or pending/in-use by a client. The updates happen in three steps:
+ *     free:
+ *     The block is free.
+ *
+ *     pending:
+ *     The block is allocated to some client and not free.
+ *
+ *     configured:
+ *     The block is allocated to some client, and assigned to that
+ *     client in MDP5_MDP_SMP_ALLOC registers.
+ *
+ *     inuse:
+ *     The block is being actively used by a client.
+ *
+ * The updates happen in the following steps:
  *
  *  1) mdp5_smp_request():
  *     When plane scanout is setup, calculate required number of
- *     blocks needed per client, and request.  Blocks not inuse or
+ *     blocks needed per client, and request. Blocks neither inuse nor
- *     pending by any other client are added to client's pending
+ *     configured nor pending by any other client are added to client's
- *     set.
+ *     pending set.
+ *     For shrinking, blocks in pending but not in configured can be freed
+ *     directly, but those already in configured will be freed later by
+ *     mdp5_smp_commit.
  *
  *  2) mdp5_smp_configure():
  *     As hw is programmed, before FLUSH, MDP5_MDP_SMP_ALLOC registers
  *     are configured for the union(pending, inuse)
+ *     Current pending is copied to configured.
+ *     It is assumed that mdp5_smp_request and mdp5_smp_configure not run
+ *     concurrently for the same pipe.
  *
  *  3) mdp5_smp_commit():
- *     After next vblank, copy pending -> inuse.  Optionally update
+ *     After next vblank, copy configured -> inuse.  Optionally update
  *     MDP5_SMP_ALLOC registers if there are newly unused blocks
  *
+ *  4) mdp5_smp_release():
+ *     Must be called after the pipe is disabled and no longer uses any SMB
+ *
  * On the next vblank after changes have been committed to hw, the
  * client's pending blocks become it's in-use blocks (and no-longer
  * in-use blocks become available to other clients).
@@ -77,6 +99,9 @@ struct mdp5_smp {
 	struct mdp5_client_smp_state client_state[MAX_CLIENTS];
 };
 
+static void update_smp_state(struct mdp5_smp *smp,
+		u32 cid, mdp5_smp_state_t *assigned);
+
 static inline
 struct mdp5_kms *get_kms(struct mdp5_smp *smp)
 {
@@ -149,7 +174,12 @@ static int smp_request_block(struct mdp5_smp *smp,
 		for (i = cur_nblks; i > nblks; i--) {
 			int blk = find_first_bit(ps->pending, cnt);
 			clear_bit(blk, ps->pending);
-			/* don't clear in global smp_state until _commit() */
+
+			/* clear in global smp_state if not in configured
+			 * otherwise until _commit()
+			 */
+			if (!test_bit(blk, ps->configured))
+				clear_bit(blk, smp->state);
 		}
 	}
 
@@ -223,10 +253,33 @@ int mdp5_smp_request(struct mdp5_smp *smp, enum mdp5_pipe pipe, u32 fmt, u32 wid
 /* Release SMP blocks for all clients of the pipe */
 void mdp5_smp_release(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 {
-	int i, nblks;
+	int i;
+	unsigned long flags;
+	int cnt = smp->blk_cnt;
+
+	for (i = 0; i < pipe2nclients(pipe); i++) {
+		mdp5_smp_state_t assigned;
+		u32 cid = pipe2client(pipe, i);
+		struct mdp5_client_smp_state *ps = &smp->client_state[cid];
+
+		spin_lock_irqsave(&smp->state_lock, flags);
+
+		/* clear hw assignment */
+		bitmap_or(assigned, ps->inuse, ps->configured, cnt);
+		update_smp_state(smp, CID_UNUSED, &assigned);
+
+		/* free to global pool */
+		bitmap_andnot(smp->state, smp->state, ps->pending, cnt);
+		bitmap_andnot(smp->state, smp->state, assigned, cnt);
+
+		/* clear client's infor */
+		bitmap_zero(ps->pending, cnt);
+		bitmap_zero(ps->configured, cnt);
+		bitmap_zero(ps->inuse, cnt);
+
+		spin_unlock_irqrestore(&smp->state_lock, flags);
+	}
 
-	for (i = 0, nblks = 0; i < pipe2nclients(pipe); i++)
-		smp_request_block(smp, pipe2client(pipe, i), 0);
 	set_fifo_thresholds(smp, pipe, 0);
 }
 
@@ -274,12 +327,20 @@ void mdp5_smp_configure(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 		u32 cid = pipe2client(pipe, i);
 		struct mdp5_client_smp_state *ps = &smp->client_state[cid];
 
-		bitmap_or(assigned, ps->inuse, ps->pending, cnt);
+		/*
+		 * if vblank has not happened since last smp_configure
+		 * skip the configure for now
+		 */
+		if (!bitmap_equal(ps->inuse, ps->configured, cnt))
+			continue;
+
+		bitmap_copy(ps->configured, ps->pending, cnt);
+		bitmap_or(assigned, ps->inuse, ps->configured, cnt);
 		update_smp_state(smp, cid, &assigned);
 	}
 }
 
-/* step #3: after vblank, copy pending -> inuse: */
+/* step #3: after vblank, copy configured -> inuse: */
 void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 {
 	int cnt = smp->blk_cnt;
@@ -295,7 +356,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 		 * using, which can be released and made available to other
 		 * clients:
 		 */
-		if (bitmap_andnot(released, ps->inuse, ps->pending, cnt)) {
+		if (bitmap_andnot(released, ps->inuse, ps->configured, cnt)) {
 			unsigned long flags;
 
 			spin_lock_irqsave(&smp->state_lock, flags);
@@ -306,7 +367,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 			update_smp_state(smp, CID_UNUSED, &released);
 		}
 
-		bitmap_copy(ps->inuse, ps->pending, cnt);
+		bitmap_copy(ps->inuse, ps->configured, cnt);
 	}
 }
 

drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h

@@ -23,6 +23,7 @@
 
 struct mdp5_client_smp_state {
 	mdp5_smp_state_t inuse;
+	mdp5_smp_state_t configured;
 	mdp5_smp_state_t pending;
 };
 

drivers/gpu/drm/msm/msm_atomic.c

@@ -283,12 +283,8 @@ int msm_atomic_commit(struct drm_device *dev,
 
 	timeout = ktime_add_ms(ktime_get(), 1000);
 
-	ret = msm_wait_fence_interruptable(dev, c->fence, &timeout);
+	/* uninterruptible wait */
-	if (ret) {
+	msm_wait_fence(dev, c->fence, &timeout, false);
-		WARN_ON(ret);  // TODO unswap state back?  or??
-		commit_destroy(c);
-		return ret;
-	}
 
 	complete_commit(c);
 

drivers/gpu/drm/msm/msm_drv.c

@@ -637,8 +637,8 @@ static void msm_debugfs_cleanup(struct drm_minor *minor)
  * Fences:
  */
 
-int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
+int msm_wait_fence(struct drm_device *dev, uint32_t fence,
-		ktime_t *timeout)
+		ktime_t *timeout , bool interruptible)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	int ret;
@@ -667,7 +667,12 @@ int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
 			remaining_jiffies = timespec_to_jiffies(&ts);
 		}
 
-		ret = wait_event_interruptible_timeout(priv->fence_event,
+		if (interruptible)
+			ret = wait_event_interruptible_timeout(priv->fence_event,
+				fence_completed(dev, fence),
+				remaining_jiffies);
+		else
+			ret = wait_event_timeout(priv->fence_event,
 				fence_completed(dev, fence),
 				remaining_jiffies);
 
@@ -853,7 +858,7 @@ static int msm_ioctl_wait_fence(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	return msm_wait_fence_interruptable(dev, args->fence, &timeout);
+	return msm_wait_fence(dev, args->fence, &timeout, true);
 }
 
 static const struct drm_ioctl_desc msm_ioctls[] = {

drivers/gpu/drm/msm/msm_drv.h

@@ -164,8 +164,8 @@ int msm_atomic_commit(struct drm_device *dev,
 
 int msm_register_mmu(struct drm_device *dev, struct msm_mmu *mmu);
 
-int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
+int msm_wait_fence(struct drm_device *dev, uint32_t fence,
-		ktime_t *timeout);
+		ktime_t *timeout, bool interruptible);
 int msm_queue_fence_cb(struct drm_device *dev,
 		struct msm_fence_cb *cb, uint32_t fence);
 void msm_update_fence(struct drm_device *dev, uint32_t fence);

drivers/gpu/drm/msm/msm_gem.c

@@ -460,7 +460,7 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout)
 		if (op & MSM_PREP_NOSYNC)
 			timeout = NULL;
 
-		ret = msm_wait_fence_interruptable(dev, fence, timeout);
+		ret = msm_wait_fence(dev, fence, timeout, true);
 	}
 
 	/* TODO cache maintenance */

drivers/gpu/drm/msm/msm_gem_prime.c

@@ -23,8 +23,12 @@
 struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
-	BUG_ON(!msm_obj->sgt);  /* should have already pinned! */
+	int npages = obj->size >> PAGE_SHIFT;
-	return msm_obj->sgt;
+
+	if (WARN_ON(!msm_obj->pages))  /* should have already pinned! */
+		return NULL;
+
+	return drm_prime_pages_to_sg(msm_obj->pages, npages);
 }
 
 void *msm_gem_prime_vmap(struct drm_gem_object *obj)

drivers/gpu/drm/nouveau/nouveau_drm.c

@@ -128,6 +128,7 @@ nouveau_cli_destroy(struct nouveau_cli *cli)
 	nvkm_vm_ref(NULL, &nvxx_client(&cli->base)->vm, NULL);
 	nvif_client_fini(&cli->base);
 	usif_client_fini(cli);
+	kfree(cli);
 }
 
 static void
@@ -865,8 +866,10 @@ nouveau_drm_preclose(struct drm_device *dev, struct drm_file *fpriv)
 
 	pm_runtime_get_sync(dev->dev);
 
+	mutex_lock(&cli->mutex);
 	if (cli->abi16)
 		nouveau_abi16_fini(cli->abi16);
+	mutex_unlock(&cli->mutex);
 
 	mutex_lock(&drm->client.mutex);
 	list_del(&cli->head);

drivers/gpu/drm/nouveau/nouveau_platform.c

@@ -92,6 +92,8 @@ static int nouveau_platform_power_down(struct nouveau_platform_gpu *gpu)
 	return 0;
 }
 
+#if IS_ENABLED(CONFIG_IOMMU_API)
+
 static void nouveau_platform_probe_iommu(struct device *dev,
 					 struct nouveau_platform_gpu *gpu)
 {
@@ -158,6 +160,20 @@ static void nouveau_platform_remove_iommu(struct device *dev,
 	}
 }
 
+#else
+
+static void nouveau_platform_probe_iommu(struct device *dev,
+					 struct nouveau_platform_gpu *gpu)
+{
+}
+
+static void nouveau_platform_remove_iommu(struct device *dev,
+					  struct nouveau_platform_gpu *gpu)
+{
+}
+
+#endif
+
 static int nouveau_platform_probe(struct platform_device *pdev)
 {
 	struct nouveau_platform_gpu *gpu;

drivers/gpu/drm/nouveau/nouveau_ttm.c

@@ -175,15 +175,24 @@ nouveau_gart_manager_new(struct ttm_mem_type_manager *man,
 	node->page_shift = 12;
 
 	switch (drm->device.info.family) {
+	case NV_DEVICE_INFO_V0_TNT:
+	case NV_DEVICE_INFO_V0_CELSIUS:
+	case NV_DEVICE_INFO_V0_KELVIN:
+	case NV_DEVICE_INFO_V0_RANKINE:
+	case NV_DEVICE_INFO_V0_CURIE:
+		break;
 	case NV_DEVICE_INFO_V0_TESLA:
 		if (drm->device.info.chipset != 0x50)
 			node->memtype = (nvbo->tile_flags & 0x7f00) >> 8;
 		break;
 	case NV_DEVICE_INFO_V0_FERMI:
 	case NV_DEVICE_INFO_V0_KEPLER:
+	case NV_DEVICE_INFO_V0_MAXWELL:
 		node->memtype = (nvbo->tile_flags & 0xff00) >> 8;
 		break;
 	default:
+		NV_WARN(drm, "%s: unhandled family type %x\n", __func__,
+			drm->device.info.family);
 		break;
 	}
 

drivers/gpu/drm/nouveau/nv04_fbcon.c

@@ -203,7 +203,7 @@ nv04_fbcon_accel_init(struct fb_info *info)
 	if (ret)
 		return ret;
 
-	if (RING_SPACE(chan, 49)) {
+	if (RING_SPACE(chan, 49 + (device->info.chipset >= 0x11 ? 4 : 0))) {
 		nouveau_fbcon_gpu_lockup(info);
 		return 0;
 	}

drivers/gpu/drm/nouveau/nv50_display.c

@@ -979,7 +979,7 @@ nv50_crtc_cursor_show_hide(struct nouveau_crtc *nv_crtc, bool show, bool update)
 {
 	struct nv50_mast *mast = nv50_mast(nv_crtc->base.dev);
 
-	if (show && nv_crtc->cursor.nvbo)
+	if (show && nv_crtc->cursor.nvbo && nv_crtc->base.enabled)
 		nv50_crtc_cursor_show(nv_crtc);
 	else
 		nv50_crtc_cursor_hide(nv_crtc);

drivers/gpu/drm/nouveau/nv50_fbcon.c

@@ -188,7 +188,7 @@ nv50_fbcon_accel_init(struct fb_info *info)
 	if (ret)
 		return ret;
 
-	ret = RING_SPACE(chan, 59);
+	ret = RING_SPACE(chan, 58);
 	if (ret) {
 		nouveau_fbcon_gpu_lockup(info);
 		return ret;
@@ -252,6 +252,7 @@ nv50_fbcon_accel_init(struct fb_info *info)
 	OUT_RING(chan, info->var.yres_virtual);
 	OUT_RING(chan, upper_32_bits(fb->vma.offset));
 	OUT_RING(chan, lower_32_bits(fb->vma.offset));
+	FIRE_RING(chan);
 
 	return 0;
 }

drivers/gpu/drm/nouveau/nvc0_fbcon.c

@@ -188,7 +188,7 @@ nvc0_fbcon_accel_init(struct fb_info *info)
 		return -EINVAL;
 	}
 
-	ret = RING_SPACE(chan, 60);
+	ret = RING_SPACE(chan, 58);
 	if (ret) {
 		WARN_ON(1);
 		nouveau_fbcon_gpu_lockup(info);

drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c

@@ -809,7 +809,7 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl,
 		case 0x00000900: type = DCB_OUTPUT_DP; mask = 2; break;
 		default:
 			nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl);
-			return 0x0000;
+			return NULL;
 		}
 	}
 

drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c

@@ -165,15 +165,31 @@ gk104_fifo_context_attach(struct nvkm_object *parent,
 	return 0;
 }
 
+static int
+gk104_fifo_chan_kick(struct gk104_fifo_chan *chan)
+{
+	struct nvkm_object *obj = (void *)chan;
+	struct gk104_fifo_priv *priv = (void *)obj->engine;
+
+	nv_wr32(priv, 0x002634, chan->base.chid);
+	if (!nv_wait(priv, 0x002634, 0x100000, 0x000000)) {
+		nv_error(priv, "channel %d [%s] kick timeout\n",
+			 chan->base.chid, nvkm_client_name(chan));
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
 static int
 gk104_fifo_context_detach(struct nvkm_object *parent, bool suspend,
 			  struct nvkm_object *object)
 {
 	struct nvkm_bar *bar = nvkm_bar(parent);
-	struct gk104_fifo_priv *priv = (void *)parent->engine;
 	struct gk104_fifo_base *base = (void *)parent->parent;
 	struct gk104_fifo_chan *chan = (void *)parent;
 	u32 addr;
+	int ret;
 
 	switch (nv_engidx(object->engine)) {
 	case NVDEV_ENGINE_SW    : return 0;
@@ -188,13 +204,9 @@ gk104_fifo_context_detach(struct nvkm_object *parent, bool suspend,
 		return -EINVAL;
 	}
 
-	nv_wr32(priv, 0x002634, chan->base.chid);
+	ret = gk104_fifo_chan_kick(chan);
-	if (!nv_wait(priv, 0x002634, 0xffffffff, chan->base.chid)) {
+	if (ret && suspend)
-		nv_error(priv, "channel %d [%s] kick timeout\n",
+		return ret;
-			 chan->base.chid, nvkm_client_name(chan));
-		if (suspend)
-			return -EBUSY;
-	}
 
 	if (addr) {
 		nv_wo32(base, addr + 0x00, 0x00000000);
@@ -319,6 +331,7 @@ gk104_fifo_chan_fini(struct nvkm_object *object, bool suspend)
 		gk104_fifo_runlist_update(priv, chan->engine);
 	}
 
+	gk104_fifo_chan_kick(chan);
 	nv_wr32(priv, 0x800000 + (chid * 8), 0x00000000);
 	return nvkm_fifo_channel_fini(&chan->base, suspend);
 }

drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c

@@ -663,6 +663,37 @@ gf100_gr_zbc_init(struct gf100_gr_priv *priv)
 		gf100_gr_zbc_clear_depth(priv, index);
 }
 
+/**
+ * Wait until GR goes idle. GR is considered idle if it is disabled by the
+ * MC (0x200) register, or GR is not busy and a context switch is not in
+ * progress.
+ */
+int
+gf100_gr_wait_idle(struct gf100_gr_priv *priv)
+{
+	unsigned long end_jiffies = jiffies + msecs_to_jiffies(2000);
+	bool gr_enabled, ctxsw_active, gr_busy;
+
+	do {
+		/*
+		 * required to make sure FIFO_ENGINE_STATUS (0x2640) is
+		 * up-to-date
+		 */
+		nv_rd32(priv, 0x400700);
+
+		gr_enabled = nv_rd32(priv, 0x200) & 0x1000;
+		ctxsw_active = nv_rd32(priv, 0x2640) & 0x8000;
+		gr_busy = nv_rd32(priv, 0x40060c) & 0x1;
+
+		if (!gr_enabled || (!gr_busy && !ctxsw_active))
+			return 0;
+	} while (time_before(jiffies, end_jiffies));
+
+	nv_error(priv, "wait for idle timeout (en: %d, ctxsw: %d, busy: %d)\n",
+		 gr_enabled, ctxsw_active, gr_busy);
+	return -EAGAIN;
+}
+
 void
 gf100_gr_mmio(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p)
 {
@@ -699,7 +730,13 @@ gf100_gr_icmd(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p)
 
 		while (addr < next) {
 			nv_wr32(priv, 0x400200, addr);
-			nv_wait(priv, 0x400700, 0x00000002, 0x00000000);
+			/**
+			 * Wait for GR to go idle after submitting a
+			 * GO_IDLE bundle
+			 */
+			if ((addr & 0xffff) == 0xe100)
+				gf100_gr_wait_idle(priv);
+			nv_wait(priv, 0x400700, 0x00000004, 0x00000000);
 			addr += init->pitch;
 		}
 	}

drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h

@@ -181,6 +181,7 @@ struct gf100_gr_oclass {
 	int ppc_nr;
 };
 
+int  gf100_gr_wait_idle(struct gf100_gr_priv *);
 void gf100_gr_mmio(struct gf100_gr_priv *, const struct gf100_gr_pack *);
 void gf100_gr_icmd(struct gf100_gr_priv *, const struct gf100_gr_pack *);
 void gf100_gr_mthd(struct gf100_gr_priv *, const struct gf100_gr_pack *);

drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c

@@ -332,9 +332,12 @@ static void
 nvkm_perfctx_dtor(struct nvkm_object *object)
 {
 	struct nvkm_pm *ppm = (void *)object->engine;
+	struct nvkm_perfctx *ctx = (void *)object;
+
 	mutex_lock(&nv_subdev(ppm)->mutex);
-	nvkm_engctx_destroy(&ppm->context->base);
+	nvkm_engctx_destroy(&ctx->base);
-	ppm->context = NULL;
+	if (ppm->context == ctx)
+		ppm->context = NULL;
 	mutex_unlock(&nv_subdev(ppm)->mutex);
 }
 
@@ -355,12 +358,11 @@ nvkm_perfctx_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 	mutex_lock(&nv_subdev(ppm)->mutex);
 	if (ppm->context == NULL)
 		ppm->context = ctx;
+	if (ctx != ppm->context)
+		ret = -EBUSY;
 	mutex_unlock(&nv_subdev(ppm)->mutex);
 
-	if (ctx != ppm->context)
+	return ret;
-		return -EBUSY;
-
-	return 0;
 }
 
 struct nvkm_oclass

drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c

@@ -1284,6 +1284,44 @@ init_zm_reg_sequence(struct nvbios_init *init)
 	}
 }
 
+/**
+ * INIT_PLL_INDIRECT - opcode 0x59
+ *
+ */
+static void
+init_pll_indirect(struct nvbios_init *init)
+{
+	struct nvkm_bios *bios = init->bios;
+	u32  reg = nv_ro32(bios, init->offset + 1);
+	u16 addr = nv_ro16(bios, init->offset + 5);
+	u32 freq = (u32)nv_ro16(bios, addr) * 1000;
+
+	trace("PLL_INDIRECT\tR[0x%06x] =PLL= VBIOS[%04x] = %dkHz\n",
+	      reg, addr, freq);
+	init->offset += 7;
+
+	init_prog_pll(init, reg, freq);
+}
+
+/**
+ * INIT_ZM_REG_INDIRECT - opcode 0x5a
+ *
+ */
+static void
+init_zm_reg_indirect(struct nvbios_init *init)
+{
+	struct nvkm_bios *bios = init->bios;
+	u32  reg = nv_ro32(bios, init->offset + 1);
+	u16 addr = nv_ro16(bios, init->offset + 5);
+	u32 data = nv_ro32(bios, addr);
+
+	trace("ZM_REG_INDIRECT\tR[0x%06x] = VBIOS[0x%04x] = 0x%08x\n",
+	      reg, addr, data);
+	init->offset += 7;
+
+	init_wr32(init, addr, data);
+}
+
 /**
  * INIT_SUB_DIRECT - opcode 0x5b
  *
@@ -2145,6 +2183,8 @@ static struct nvbios_init_opcode {
 	[0x56] = { init_condition_time },
 	[0x57] = { init_ltime },
 	[0x58] = { init_zm_reg_sequence },
+	[0x59] = { init_pll_indirect },
+	[0x5a] = { init_zm_reg_indirect },
 	[0x5b] = { init_sub_direct },
 	[0x5c] = { init_jump },
 	[0x5e] = { init_i2c_if },

drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c

@@ -180,7 +180,8 @@ gt215_clk_info(struct nvkm_clk *clock, int clk, u32 khz,
 	       struct gt215_clk_info *info)
 {
 	struct gt215_clk_priv *priv = (void *)clock;
-	u32 oclk, sclk, sdiv, diff;
+	u32 oclk, sclk, sdiv;
+	s32 diff;
 
 	info->clk = 0;
 

drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c

@@ -38,6 +38,14 @@ gk20a_ibus_init_priv_ring(struct gk20a_ibus_priv *priv)
 	nv_wr32(priv, 0x12004c, 0x4);
 	nv_wr32(priv, 0x122204, 0x2);
 	nv_rd32(priv, 0x122204);
+
+	/*
+	 * Bug: increase clock timeout to avoid operation failure at high
+	 * gpcclk rate.
+	 */
+	nv_wr32(priv, 0x122354, 0x800);
+	nv_wr32(priv, 0x128328, 0x800);
+	nv_wr32(priv, 0x124320, 0x800);
 }
 
 static void

drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c

@@ -50,7 +50,12 @@ nv04_instobj_dtor(struct nvkm_object *object)
 {
 	struct nv04_instmem_priv *priv = (void *)nvkm_instmem(object);
 	struct nv04_instobj_priv *node = (void *)object;
+	struct nvkm_subdev *subdev = (void *)priv;
+
+	mutex_lock(&subdev->mutex);
 	nvkm_mm_free(&priv->heap, &node->mem);
+	mutex_unlock(&subdev->mutex);
+
 	nvkm_instobj_destroy(&node->base);
 }
 
@@ -62,6 +67,7 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 	struct nv04_instmem_priv *priv = (void *)nvkm_instmem(parent);
 	struct nv04_instobj_priv *node;
 	struct nvkm_instobj_args *args = data;
+	struct nvkm_subdev *subdev = (void *)priv;
 	int ret;
 
 	if (!args->align)
@@ -72,8 +78,10 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 	if (ret)
 		return ret;
 
+	mutex_lock(&subdev->mutex);
 	ret = nvkm_mm_head(&priv->heap, 0, 1, args->size, args->size,
 			   args->align, &node->mem);
+	mutex_unlock(&subdev->mutex);
 	if (ret)
 		return ret;
 

drivers/gpu/drm/radeon/atombios_encoders.c

@@ -2299,8 +2299,7 @@ radeon_atom_encoder_mode_set(struct drm_encoder *encoder,
 	encoder_mode = atombios_get_encoder_mode(encoder);
 	if (connector && (radeon_audio != 0) &&
 	    ((encoder_mode == ATOM_ENCODER_MODE_HDMI) ||
-	     (ENCODER_MODE_IS_DP(encoder_mode) &&
+	     ENCODER_MODE_IS_DP(encoder_mode)))
-	      drm_detect_monitor_audio(radeon_connector_edid(connector)))))
 		radeon_audio_mode_set(encoder, adjusted_mode);
 }
 

drivers/gpu/drm/radeon/dce6_afmt.c

@@ -93,30 +93,26 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder)
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 offset;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->offset;
+	WREG32(AFMT_AUDIO_SRC_CONTROL +  dig->afmt->offset,
-
+	       AFMT_AUDIO_SRC_SELECT(dig->pin->id));
-	WREG32(AFMT_AUDIO_SRC_CONTROL + offset,
-	       AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id));
 }
 
 void dce6_afmt_write_latency_fields(struct drm_encoder *encoder,
-		struct drm_connector *connector, struct drm_display_mode *mode)
+				    struct drm_connector *connector,
+				    struct drm_display_mode *mode)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 tmp = 0, offset;
+	u32 tmp = 0;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
 		if (connector->latency_present[1])
 			tmp = VIDEO_LIPSYNC(connector->video_latency[1]) |
@@ -130,24 +126,24 @@ void dce6_afmt_write_latency_fields(struct drm_encoder *encoder,
 		else
 			tmp = VIDEO_LIPSYNC(0) | AUDIO_LIPSYNC(0);
 	}
-	WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp);
+	WREG32_ENDPOINT(dig->pin->offset,
+			AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp);
 }
 
 void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder,
-	u8 *sadb, int sad_count)
+					     u8 *sadb, int sad_count)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 offset, tmp;
+	u32 tmp;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	/* program the speaker allocation */
-	tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
+	tmp = RREG32_ENDPOINT(dig->pin->offset,
+			      AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
 	tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK);
 	/* set HDMI mode */
 	tmp |= HDMI_CONNECTION;
@@ -155,24 +151,24 @@ void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder,
 		tmp |= SPEAKER_ALLOCATION(sadb[0]);
 	else
 		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
-	WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
+	WREG32_ENDPOINT(dig->pin->offset,
+			AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
 }
 
 void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder,
-	u8 *sadb, int sad_count)
+					   u8 *sadb, int sad_count)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 offset, tmp;
+	u32 tmp;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	/* program the speaker allocation */
-	tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
+	tmp = RREG32_ENDPOINT(dig->pin->offset,
+			      AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
 	tmp &= ~(HDMI_CONNECTION | SPEAKER_ALLOCATION_MASK);
 	/* set DP mode */
 	tmp |= DP_CONNECTION;
@@ -180,13 +176,13 @@ void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder,
 		tmp |= SPEAKER_ALLOCATION(sadb[0]);
 	else
 		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
-	WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
+	WREG32_ENDPOINT(dig->pin->offset,
+			AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
 }
 
 void dce6_afmt_write_sad_regs(struct drm_encoder *encoder,
-	struct cea_sad *sads, int sad_count)
+			      struct cea_sad *sads, int sad_count)
 {
-	u32 offset;
 	int i;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
@@ -206,11 +202,9 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder,
 		{ AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
 	};
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
 		u32 value = 0;
 		u8 stereo_freqs = 0;
@@ -237,7 +231,7 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder,
 
 		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
 
-		WREG32_ENDPOINT(offset, eld_reg_to_type[i][0], value);
+		WREG32_ENDPOINT(dig->pin->offset, eld_reg_to_type[i][0], value);
 	}
 }
 
@@ -253,7 +247,7 @@ void dce6_audio_enable(struct radeon_device *rdev,
 }
 
 void dce6_hdmi_audio_set_dto(struct radeon_device *rdev,
-	struct radeon_crtc *crtc, unsigned int clock)
+			     struct radeon_crtc *crtc, unsigned int clock)
 {
 	/* Two dtos; generally use dto0 for HDMI */
 	u32 value = 0;
@@ -272,7 +266,7 @@ void dce6_hdmi_audio_set_dto(struct radeon_device *rdev,
 }
 
 void dce6_dp_audio_set_dto(struct radeon_device *rdev,
-	struct radeon_crtc *crtc, unsigned int clock)
+			   struct radeon_crtc *crtc, unsigned int clock)
 {
 	/* Two dtos; generally use dto1 for DP */
 	u32 value = 0;

drivers/gpu/drm/radeon/radeon_audio.c

@@ -245,6 +245,28 @@ static struct radeon_audio_funcs dce6_dp_funcs = {
 static void radeon_audio_enable(struct radeon_device *rdev,
 				struct r600_audio_pin *pin, u8 enable_mask)
 {
+	struct drm_encoder *encoder;
+	struct radeon_encoder *radeon_encoder;
+	struct radeon_encoder_atom_dig *dig;
+	int pin_count = 0;
+
+	if (!pin)
+		return;
+
+	if (rdev->mode_info.mode_config_initialized) {
+		list_for_each_entry(encoder, &rdev->ddev->mode_config.encoder_list, head) {
+			if (radeon_encoder_is_digital(encoder)) {
+				radeon_encoder = to_radeon_encoder(encoder);
+				dig = radeon_encoder->enc_priv;
+				if (dig->pin == pin)
+					pin_count++;
+			}
+		}
+
+		if ((pin_count > 1) && (enable_mask == 0))
+			return;
+	}
+
 	if (rdev->audio.funcs->enable)
 		rdev->audio.funcs->enable(rdev, pin, enable_mask);
 }
@@ -336,24 +358,13 @@ void radeon_audio_endpoint_wreg(struct radeon_device *rdev, u32 offset,
 
 static void radeon_audio_write_sad_regs(struct drm_encoder *encoder)
 {
-	struct radeon_encoder *radeon_encoder;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
-	struct drm_connector *connector;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
-	struct radeon_connector *radeon_connector = NULL;
 	struct cea_sad *sads;
 	int sad_count;
 
-	list_for_each_entry(connector,
+	if (!connector)
-		&encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
 		return;
-	}
 
 	sad_count = drm_edid_to_sad(radeon_connector_edid(connector), &sads);
 	if (sad_count <= 0) {
@@ -362,8 +373,6 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder)
 	}
 	BUG_ON(!sads);
 
-	radeon_encoder = to_radeon_encoder(encoder);
-
 	if (radeon_encoder->audio && radeon_encoder->audio->write_sad_regs)
 		radeon_encoder->audio->write_sad_regs(encoder, sads, sad_count);
 
@@ -372,27 +381,16 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder)
 
 static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder)
 {
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
 	u8 *sadb = NULL;
 	int sad_count;
 
-	list_for_each_entry(connector,
+	if (!connector)
-			    &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
 		return;
-	}
 
-	sad_count = drm_edid_to_speaker_allocation(
+	sad_count = drm_edid_to_speaker_allocation(radeon_connector_edid(connector),
-		radeon_connector_edid(connector), &sadb);
+						   &sadb);
 	if (sad_count < 0) {
 		DRM_DEBUG("Couldn't read Speaker Allocation Data Block: %d\n",
 			  sad_count);
@@ -406,26 +404,13 @@ static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder)
 }
 
 static void radeon_audio_write_latency_fields(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+					      struct drm_display_mode *mode)
 {
-	struct radeon_encoder *radeon_encoder;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
-	struct drm_connector *connector;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
-	struct radeon_connector *radeon_connector = 0;
 
-	list_for_each_entry(connector,
+	if (!connector)
-		&encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
 		return;
-	}
-
-	radeon_encoder = to_radeon_encoder(encoder);
 
 	if (radeon_encoder->audio && radeon_encoder->audio->write_latency_fields)
 		radeon_encoder->audio->write_latency_fields(encoder, connector, mode);
@@ -451,29 +436,23 @@ static void radeon_audio_select_pin(struct drm_encoder *encoder)
 }
 
 void radeon_audio_detect(struct drm_connector *connector,
+			 struct drm_encoder *encoder,
 			 enum drm_connector_status status)
 {
-	struct radeon_device *rdev;
+	struct drm_device *dev = connector->dev;
-	struct radeon_encoder *radeon_encoder;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig;
 
-	if (!connector || !connector->encoder)
-		return;
-
-	rdev = connector->encoder->dev->dev_private;
-
 	if (!radeon_audio_chipset_supported(rdev))
 		return;
 
-	radeon_encoder = to_radeon_encoder(connector->encoder);
+	if (!radeon_encoder_is_digital(encoder))
+		return;
+
 	dig = radeon_encoder->enc_priv;
 
 	if (status == connector_status_connected) {
-		if (!drm_detect_monitor_audio(radeon_connector_edid(connector))) {
-			radeon_encoder->audio = NULL;
-			return;
-		}
-
 		if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) {
 			struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 
@@ -486,11 +465,17 @@ void radeon_audio_detect(struct drm_connector *connector,
 			radeon_encoder->audio = rdev->audio.hdmi_funcs;
 		}
 
-		dig->afmt->pin = radeon_audio_get_pin(connector->encoder);
+		if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
-		radeon_audio_enable(rdev, dig->afmt->pin, 0xf);
+			if (!dig->pin)
+				dig->pin = radeon_audio_get_pin(encoder);
+			radeon_audio_enable(rdev, dig->pin, 0xf);
+		} else {
+			radeon_audio_enable(rdev, dig->pin, 0);
+			dig->pin = NULL;
+		}
 	} else {
-		radeon_audio_enable(rdev, dig->afmt->pin, 0);
+		radeon_audio_enable(rdev, dig->pin, 0);
-		dig->afmt->pin = NULL;
+		dig->pin = NULL;
 	}
 }
 
@@ -518,29 +503,18 @@ static void radeon_audio_set_dto(struct drm_encoder *encoder, unsigned int clock
 }
 
 static int radeon_audio_set_avi_packet(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+				       struct drm_display_mode *mode)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	struct drm_connector *connector;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
-	struct radeon_connector *radeon_connector = NULL;
 	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
 	struct hdmi_avi_infoframe frame;
 	int err;
 
-	list_for_each_entry(connector,
+	if (!connector)
-		&encoder->dev->mode_config.connector_list, head) {
+		return -EINVAL;
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return -ENOENT;
-	}
 
 	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
 	if (err < 0) {
@@ -563,8 +537,8 @@ static int radeon_audio_set_avi_packet(struct drm_encoder *encoder,
 		return err;
 	}
 
-	if (dig && dig->afmt &&
+	if (dig && dig->afmt && radeon_encoder->audio &&
-		radeon_encoder->audio && radeon_encoder->audio->set_avi_packet)
+	    radeon_encoder->audio->set_avi_packet)
 		radeon_encoder->audio->set_avi_packet(rdev, dig->afmt->offset,
 			buffer, sizeof(buffer));
 
@@ -722,30 +696,41 @@ static void radeon_audio_hdmi_mode_set(struct drm_encoder *encoder,
 {
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 
 	if (!dig || !dig->afmt)
 		return;
 
-	radeon_audio_set_mute(encoder, true);
+	if (!connector)
-
-	radeon_audio_write_speaker_allocation(encoder);
-	radeon_audio_write_sad_regs(encoder);
-	radeon_audio_write_latency_fields(encoder, mode);
-	radeon_audio_set_dto(encoder, mode->clock);
-	radeon_audio_set_vbi_packet(encoder);
-	radeon_hdmi_set_color_depth(encoder);
-	radeon_audio_update_acr(encoder, mode->clock);
-	radeon_audio_set_audio_packet(encoder);
-	radeon_audio_select_pin(encoder);
-
-	if (radeon_audio_set_avi_packet(encoder, mode) < 0)
 		return;
 
-	radeon_audio_set_mute(encoder, false);
+	if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
+		radeon_audio_set_mute(encoder, true);
+
+		radeon_audio_write_speaker_allocation(encoder);
+		radeon_audio_write_sad_regs(encoder);
+		radeon_audio_write_latency_fields(encoder, mode);
+		radeon_audio_set_dto(encoder, mode->clock);
+		radeon_audio_set_vbi_packet(encoder);
+		radeon_hdmi_set_color_depth(encoder);
+		radeon_audio_update_acr(encoder, mode->clock);
+		radeon_audio_set_audio_packet(encoder);
+		radeon_audio_select_pin(encoder);
+
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
+
+		radeon_audio_set_mute(encoder, false);
+	} else {
+		radeon_hdmi_set_color_depth(encoder);
+
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
+	}
 }
 
 static void radeon_audio_dp_mode_set(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+				     struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -759,22 +744,27 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder,
 	if (!dig || !dig->afmt)
 		return;
 
-	radeon_audio_write_speaker_allocation(encoder);
+	if (!connector)
-	radeon_audio_write_sad_regs(encoder);
-	radeon_audio_write_latency_fields(encoder, mode);
-	if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev))
-		radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10);
-	else
-		radeon_audio_set_dto(encoder, dig_connector->dp_clock);
-	radeon_audio_set_audio_packet(encoder);
-	radeon_audio_select_pin(encoder);
-
-	if (radeon_audio_set_avi_packet(encoder, mode) < 0)
 		return;
+
+	if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
+		radeon_audio_write_speaker_allocation(encoder);
+		radeon_audio_write_sad_regs(encoder);
+		radeon_audio_write_latency_fields(encoder, mode);
+		if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev))
+			radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10);
+		else
+			radeon_audio_set_dto(encoder, dig_connector->dp_clock);
+		radeon_audio_set_audio_packet(encoder);
+		radeon_audio_select_pin(encoder);
+
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
+	}
 }
 
 void radeon_audio_mode_set(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+			   struct drm_display_mode *mode)
 {
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);