Merge tag 'drm-intel-next-fixes-2015-09-02' into drm-intel-next-queued

Backmerge -fixes since there's more DDI-E related cleanups on top of the pile of -fixes for skl that just landed for 4.3. Conflicts: drivers/gpu/drm/i915/intel_display.c drivers/gpu/drm/i914/intel_dp.c drivers/gpu/drm/i915/intel_lrc.c Conflicts are all fairly harmless adjacent line stuff. Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
2015-09-02 14:33:42 +02:00 · 2015-09-02 14:33:42 +02:00 · e93c28f393
parent 85a62bf9d8 6fa2d19793
commit e93c28f393
627 changed files with 48288 additions and 17224 deletions
--- a/.get_maintainer.ignore
+++ b/.get_maintainer.ignore
@ -0,0 +1 @@
+Christoph Hellwig <hch@lst.de>
--- a/.mailmap
+++ b/.mailmap
@ -17,6 +17,7 @@ Aleksey Gorelov <aleksey_gorelov@phoenix.com>
 Al Viro <viro@ftp.linux.org.uk>
 Al Viro <viro@zenIV.linux.org.uk>
 Andreas Herrmann <aherrman@de.ibm.com>
+Andrey Ryabinin <ryabinin.a.a@gmail.com> <a.ryabinin@samsung.com>
 Andrew Morton <akpm@linux-foundation.org>
 Andrew Vasquez <andrew.vasquez@qlogic.com>
 Andy Adamson <andros@citi.umich.edu>
--- a/Documentation/devicetree/bindings/arm/cpus.txt
+++ b/Documentation/devicetree/bindings/arm/cpus.txt
@ -199,6 +199,7 @@ nodes to be present and contain the properties described below.
 			    "qcom,kpss-acc-v1"
 			    "qcom,kpss-acc-v2"
 			    "rockchip,rk3066-smp"
+			    "ste,dbx500-smp"

 	- cpu-release-addr
 		Usage: required for systems that have an "enable-method"
--- a/Documentation/devicetree/bindings/drm/msm/dsi.txt
+++ b/Documentation/devicetree/bindings/drm/msm/dsi.txt
@ -30,20 +30,27 @@ Optional properties:
 - panel@0: Node of panel connected to this DSI controller.
  See files in Documentation/devicetree/bindings/panel/ for each supported
  panel.
- qcom,dual-panel-mode: Boolean value indicating if the DSI controller is
+- qcom,dual-dsi-mode: Boolean value indicating if the DSI controller is
  driving a panel which needs 2 DSI links.
- qcom,master-panel: Boolean value indicating if the DSI controller is driving
+- qcom,master-dsi: Boolean value indicating if the DSI controller is driving
  the master link of the 2-DSI panel.
- qcom,sync-dual-panel: Boolean value indicating if the DSI controller is
+- qcom,sync-dual-dsi: Boolean value indicating if the DSI controller is
  driving a 2-DSI panel whose 2 links need receive command simultaneously.
 - interrupt-parent: phandle to the MDP block if the interrupt signal is routed
  through MDP block
+- pinctrl-names: the pin control state names; should contain "default"
+- pinctrl-0: the default pinctrl state (active)
+- pinctrl-n: the "sleep" pinctrl state
+- port: DSI controller output port. This contains one endpoint subnode, with its
+  remote-endpoint set to the phandle of the connected panel's endpoint.
+  See Documentation/devicetree/bindings/graph.txt for device graph info.

 DSI PHY:
 Required properties:
 - compatible: Could be the following
  * "qcom,dsi-phy-28nm-hpm"
  * "qcom,dsi-phy-28nm-lp"
+  * "qcom,dsi-phy-20nm"
 - reg: Physical base address and length of the registers of PLL, PHY and PHY
  regulator
 - reg-names: The names of register regions. The following regions are required:
@ -59,6 +66,10 @@ Required properties:
  * "iface_clk"
 - vddio-supply: phandle to vdd-io regulator device node

+Optional properties:
+- qcom,dsi-phy-regulator-ldo-mode: Boolean value indicating if the LDO mode PHY
+  regulator is wanted.
+
 Example:
 	mdss_dsi0: qcom,mdss_dsi@fd922800 {
 		compatible = "qcom,mdss-dsi-ctrl";
@ -90,9 +101,13 @@ Example:

 		qcom,dsi-phy = <&mdss_dsi_phy0>;

-		qcom,dual-panel-mode;
-		qcom,master-panel;
-		qcom,sync-dual-panel;
+		qcom,dual-dsi-mode;
+		qcom,master-dsi;
+		qcom,sync-dual-dsi;
+
+		pinctrl-names = "default", "sleep";
+		pinctrl-0 = <&mdss_dsi_active>;
+		pinctrl-1 = <&mdss_dsi_suspend>;

 		panel: panel@0 {
 			compatible = "sharp,lq101r1sx01";
@ -101,6 +116,18 @@ Example:

 			power-supply = <...>;
 			backlight = <...>;
+
+			port {
+				panel_in: endpoint {
+					remote-endpoint = <&dsi0_out>;
+				};
+			};
+		};
+
+		port {
+			dsi0_out: endpoint {
+				remote-endpoint = <&panel_in>;
+			};
 		};
 	};

@ -117,4 +144,6 @@ Example:
 		clock-names = "iface_clk";
 		clocks = <&mmcc MDSS_AHB_CLK>;
 		vddio-supply = <&pma8084_l12>;
+
+		qcom,dsi-phy-regulator-ldo-mode;
 	};
--- a/Documentation/devicetree/bindings/drm/msm/hdmi.txt
+++ b/Documentation/devicetree/bindings/drm/msm/hdmi.txt
@ -2,8 +2,9 @@ Qualcomm adreno/snapdragon hdmi output

 Required properties:
 - compatible: one of the following
+   * "qcom,hdmi-tx-8994"
   * "qcom,hdmi-tx-8084"
-   * "qcom,hdmi-tx-8074"
+   * "qcom,hdmi-tx-8974"
   * "qcom,hdmi-tx-8660"
   * "qcom,hdmi-tx-8960"
 - reg: Physical base address and length of the controller's registers
--- a/Documentation/devicetree/bindings/gpu/nvidia,tegra20-host1x.txt
+++ b/Documentation/devicetree/bindings/gpu/nvidia,tegra20-host1x.txt
@ -197,9 +197,11 @@ of the following host1x client modules:
 - sor: serial output resource

  Required properties:
-  - compatible: For Tegra124, must contain "nvidia,tegra124-sor".  Otherwise,
-    must contain '"nvidia,<chip>-sor", "nvidia,tegra124-sor"', where <chip>
-    is tegra132.
+  - compatible: Should be:
+    - "nvidia,tegra124-sor": for Tegra124 and Tegra132
+    - "nvidia,tegra132-sor": for Tegra132
+    - "nvidia,tegra210-sor": for Tegra210
+    - "nvidia,tegra210-sor1": for Tegra210
  - reg: Physical base address and length of the controller's registers.
  - interrupts: The interrupt outputs from the controller.
  - clocks: Must contain an entry for each entry in clock-names.
--- a/Documentation/devicetree/bindings/gpu/st,stih4xx.txt
+++ b/Documentation/devicetree/bindings/gpu/st,stih4xx.txt
@ -52,10 +52,9 @@ STMicroelectronics stih4xx platforms
    See ../reset/reset.txt for details.
  - reset-names: names of the resets listed in resets property in the same
    order.
-  - ranges: to allow probing of subdevices

 - sti-hdmi: hdmi output block
-  must be a child of sti-tvout
+  must be a child of sti-display-subsystem
  Required properties:
  - compatible: "st,stih<chip>-hdmi";
  - reg: Physical base address of the IP registers and length of memory mapped region.
@ -72,7 +71,7 @@ STMicroelectronics stih4xx platforms

 sti-hda:
  Required properties:
-  must be a child of sti-tvout
+  must be a child of sti-display-subsystem
  - compatible: "st,stih<chip>-hda"
  - reg: Physical base address of the IP registers and length of memory mapped region.
  - reg-names: names of the mapped memory regions listed in regs property in
@ -85,7 +84,7 @@ sti-hda:

 sti-dvo:
  Required properties:
-  must be a child of sti-tvout
+  must be a child of sti-display-subsystem
  - compatible: "st,stih<chip>-dvo"
  - reg: Physical base address of the IP registers and length of memory mapped region.
  - reg-names: names of the mapped memory regions listed in regs property in
@ -195,38 +194,37 @@ Example:
 			reg-names	= "tvout-reg", "hda-reg", "syscfg";
 			reset-names     = "tvout";
 			resets          = <&softreset STIH416_HDTVOUT_SOFTRESET>;
-			ranges;
+		};

-			sti-hdmi@fe85c000 {
-				compatible	= "st,stih416-hdmi";
-				reg		= <0xfe85c000 0x1000>, <0xfe830000 0x10000>;
-				reg-names	= "hdmi-reg", "syscfg";
-				interrupts	= <GIC_SPI 173 IRQ_TYPE_NONE>;
-				interrupt-names	= "irq";
-				clock-names	= "pix", "tmds", "phy", "audio";
-				clocks          = <&clockgen_c_vcc CLK_S_PIX_HDMI>, <&clockgen_c_vcc CLK_S_TMDS_HDMI>, <&clockgen_c_vcc CLK_S_HDMI_REJECT_PLL>, <&clockgen_b1 CLK_S_PCM_0>;
-			};
+		sti-hdmi@fe85c000 {
+			compatible	= "st,stih416-hdmi";
+			reg		= <0xfe85c000 0x1000>, <0xfe830000 0x10000>;
+			reg-names	= "hdmi-reg", "syscfg";
+			interrupts	= <GIC_SPI 173 IRQ_TYPE_NONE>;
+			interrupt-names	= "irq";
+			clock-names	= "pix", "tmds", "phy", "audio";
+			clocks          = <&clockgen_c_vcc CLK_S_PIX_HDMI>, <&clockgen_c_vcc CLK_S_TMDS_HDMI>, <&clockgen_c_vcc CLK_S_HDMI_REJECT_PLL>, <&clockgen_b1 CLK_S_PCM_0>;
+		};

-			sti-hda@fe85a000 {
-				compatible	= "st,stih416-hda";
-				reg		= <0xfe85a000 0x400>, <0xfe83085c 0x4>;
-				reg-names	= "hda-reg", "video-dacs-ctrl";
-				clock-names	= "pix", "hddac";
-				clocks          = <&clockgen_c_vcc CLK_S_PIX_HD>, <&clockgen_c_vcc CLK_S_HDDAC>;
-			};
+		sti-hda@fe85a000 {
+			compatible	= "st,stih416-hda";
+			reg		= <0xfe85a000 0x400>, <0xfe83085c 0x4>;
+			reg-names	= "hda-reg", "video-dacs-ctrl";
+			clock-names	= "pix", "hddac";
+			clocks          = <&clockgen_c_vcc CLK_S_PIX_HD>, <&clockgen_c_vcc CLK_S_HDDAC>;
+		};

-			sti-dvo@8d00400 {
-				compatible	= "st,stih407-dvo";
-				reg		= <0x8d00400 0x200>;
-				reg-names	= "dvo-reg";
-				clock-names	= "dvo_pix", "dvo",
-						  "main_parent", "aux_parent";
-				clocks		= <&clk_s_d2_flexgen CLK_PIX_DVO>, <&clk_s_d2_flexgen CLK_DVO>,
-						  <&clk_s_d2_quadfs 0>, <&clk_s_d2_quadfs 1>;
-				pinctrl-names	= "default";
-				pinctrl-0	= <&pinctrl_dvo>;
-				sti,panel	= <&panel_dvo>;
-			};
+		sti-dvo@8d00400 {
+			compatible	= "st,stih407-dvo";
+			reg		= <0x8d00400 0x200>;
+			reg-names	= "dvo-reg";
+			clock-names	= "dvo_pix", "dvo",
+					  "main_parent", "aux_parent";
+			clocks		= <&clk_s_d2_flexgen CLK_PIX_DVO>, <&clk_s_d2_flexgen CLK_DVO>,
+					  <&clk_s_d2_quadfs 0>, <&clk_s_d2_quadfs 1>;
+			pinctrl-names	= "default";
+			pinctrl-0	= <&pinctrl_dvo>;
+			sti,panel	= <&panel_dvo>;
 		};

 		sti-hqvdp@9c000000 {
@ -237,7 +235,7 @@ Example:
 				reset-names     = "hqvdp";
 				resets          = <&softreset STIH407_HDQVDP_SOFTRESET>;
 				st,vtg		= <&vtg_main>;
-			};
+		};
 	};
 	...
 };
--- a/Documentation/devicetree/bindings/panel/auo,b080uan01.txt
+++ b/Documentation/devicetree/bindings/panel/auo,b080uan01.txt
@ -0,0 +1,7 @@
+AU Optronics Corporation 8.0" WUXGA TFT LCD panel
+
+Required properties:
+- compatible: should be "auo,b101ean01"
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
--- a/Documentation/devicetree/bindings/panel/lg,lg4573.txt
+++ b/Documentation/devicetree/bindings/panel/lg,lg4573.txt
@ -0,0 +1,19 @@
+LG LG4573 TFT Liquid Crystal Display with SPI control bus
+
+Required properties:
+  - compatible: "lg,lg4573"
+  - reg: address of the panel on the SPI bus
+
+The panel must obey rules for SPI slave device specified in document [1].
+
+[1]: Documentation/devicetree/bindings/spi/spi-bus.txt
+
+Example:
+
+	lcd_panel: display@0 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "lg,lg4573";
+		spi-max-frequency = <10000000>;
+		reg = <0>;
+	};
--- a/Documentation/devicetree/bindings/panel/nec,nl4827hc19-05b.txt
+++ b/Documentation/devicetree/bindings/panel/nec,nl4827hc19-05b.txt
@ -0,0 +1,7 @@
+NEC LCD Technologies,Ltd. WQVGA TFT LCD panel
+
+Required properties:
+- compatible: should be "nec,nl4827hc19-05b"
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
--- a/Documentation/devicetree/bindings/panel/okaya,rs800480t-7x0gp.txt
+++ b/Documentation/devicetree/bindings/panel/okaya,rs800480t-7x0gp.txt
@ -0,0 +1,7 @@
+OKAYA Electric America, Inc. RS800480T-7X0GP 7" WVGA LCD panel
+
+Required properties:
+- compatible: should be "okaya,rs800480t-7x0gp"
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
--- a/Documentation/devicetree/bindings/vendor-prefixes.txt
+++ b/Documentation/devicetree/bindings/vendor-prefixes.txt
@ -140,6 +140,7 @@ mundoreader	Mundo Reader S.L.
 murata	Murata Manufacturing Co., Ltd.
 mxicy	Macronix International Co., Ltd.
 national	National Semiconductor
+nec	NEC LCD Technologies, Ltd.
 neonode		Neonode Inc.
 netgear	NETGEAR
 netlogic	Broadcom Corporation (formerly NetLogic Microsystems)
@ -148,6 +149,7 @@ nintendo	Nintendo
 nokia	Nokia
 nvidia	NVIDIA
 nxp	NXP Semiconductors
+okaya	Okaya Electric America, Inc.
 onnn	ON Semiconductor Corp.
 opencores	OpenCores.org
 ortustech	Ortus Technology Co., Ltd.
--- a/Documentation/devicetree/bindings/video/fsl,dcu.txt
+++ b/Documentation/devicetree/bindings/video/fsl,dcu.txt
@ -0,0 +1,22 @@
+Device Tree bindings for Freescale DCU DRM Driver
+
+Required properties:
+- compatible:		Should be one of
+	* "fsl,ls1021a-dcu".
+	* "fsl,vf610-dcu".
+
+- reg:			Address and length of the register set for dcu.
+- clocks:		From common clock binding: handle to dcu clock.
+- clock-names:		From common clock binding: Shall be "dcu".
+- big-endian		Boolean property, LS1021A DCU registers are big-endian.
+- fsl,panel:		The phandle to panel node.
+
+Examples:
+dcu: dcu@2ce0000 {
+	compatible = "fsl,ls1021a-dcu";
+	reg = <0x0 0x2ce0000 0x0 0x10000>;
+	clocks = <&platform_clk 0>;
+	clock-names = "dcu";
+	big-endian;
+	fsl,panel = <&panel>;
+};
--- a/18
+++ b/18
@ -3555,6 +3555,15 @@ F:	drivers/gpu/drm/exynos/
 F:	include/drm/exynos*
 F:	include/uapi/drm/exynos*

+DRM DRIVERS FOR FREESCALE DCU
+M:	Jianwei Wang <jianwei.wang.chn@gmail.com>
+M:	Alison Wang <alison.wang@freescale.com>
+L:	dri-devel@lists.freedesktop.org
+S:	Supported
+F:	drivers/gpu/drm/fsl-dcu/
+F:	Documentation/devicetree/bindings/video/fsl,dcu.txt
+F:	Documentation/devicetree/bindings/panel/nec,nl4827hc19_05b.txt
+
 DRM DRIVERS FOR FREESCALE IMX
 M:	Philipp Zabel <p.zabel@pengutronix.de>
 L:	dri-devel@lists.freedesktop.org
@ -3592,6 +3601,15 @@ S:	Maintained
 F:	drivers/gpu/drm/rockchip/
 F:	Documentation/devicetree/bindings/video/rockchip*

+DRM DRIVERS FOR STI
+M:	Benjamin Gaignard <benjamin.gaignard@linaro.org>
+M:	Vincent Abriou <vincent.abriou@st.com>
+L:	dri-devel@lists.freedesktop.org
+T:	git http://git.linaro.org/people/benjamin.gaignard/kernel.git
+S:	Maintained
+F:	drivers/gpu/drm/sti
+F:	Documentation/devicetree/bindings/gpu/st,stih4xx.txt
+
 DSBR100 USB FM RADIO DRIVER
 M:	Alexey Klimov <klimov.linux@gmail.com>
 L:	linux-media@vger.kernel.org
--- a/2
+++ b/2
@ -1,7 +1,7 @@
 VERSION = 4
 PATCHLEVEL = 2
 SUBLEVEL = 0
-EXTRAVERSION = -rc6
+EXTRAVERSION = -rc8
 NAME = Hurr durr I'ma sheep

 # *DOCUMENTATION*
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@ -312,6 +312,9 @@ INSTALL_TARGETS	= zinstall uinstall install

 PHONY += bzImage $(BOOT_TARGETS) $(INSTALL_TARGETS)

+bootpImage uImage: zImage
+zImage: Image
+
 $(BOOT_TARGETS): vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/$@

--- a/arch/arm/boot/dts/dra7.dtsi
+++ b/arch/arm/boot/dts/dra7.dtsi
@ -116,7 +116,7 @@ scm: scm@2000 {
 				ranges = <0 0x2000 0x2000>;

 				scm_conf: scm_conf@0 {
-					compatible = "syscon";
+					compatible = "syscon", "simple-bus";
 					reg = <0x0 0x1400>;
 					#address-cells = <1>;
 					#size-cells = <1>;
--- a/arch/arm/boot/dts/imx6qdl.dtsi
+++ b/arch/arm/boot/dts/imx6qdl.dtsi
@ -181,10 +181,10 @@ pcie: pcie@0x01000000 {
 			interrupt-names = "msi";
 			#interrupt-cells = <1>;
 			interrupt-map-mask = <0 0 0 0x7>;
-			interrupt-map = <0 0 0 1 &intc GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>,
-			                <0 0 0 2 &intc GIC_SPI 122 IRQ_TYPE_LEVEL_HIGH>,
-			                <0 0 0 3 &intc GIC_SPI 121 IRQ_TYPE_LEVEL_HIGH>,
-			                <0 0 0 4 &intc GIC_SPI 120 IRQ_TYPE_LEVEL_HIGH>;
+			interrupt-map = <0 0 0 1 &gpc GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>,
+			                <0 0 0 2 &gpc GIC_SPI 122 IRQ_TYPE_LEVEL_HIGH>,
+			                <0 0 0 3 &gpc GIC_SPI 121 IRQ_TYPE_LEVEL_HIGH>,
+			                <0 0 0 4 &gpc GIC_SPI 120 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&clks IMX6QDL_CLK_PCIE_AXI>,
 				 <&clks IMX6QDL_CLK_LVDS1_GATE>,
 				 <&clks IMX6QDL_CLK_PCIE_REF_125M>;
--- a/arch/arm/boot/dts/k2e.dtsi
+++ b/arch/arm/boot/dts/k2e.dtsi
@ -131,10 +131,17 @@ pcie_intc1: legacy-interrupt-controller {
 					<GIC_SPI 376 IRQ_TYPE_EDGE_RISING>;
 			};
 		};
+
+		mdio: mdio@24200f00 {
+			compatible	= "ti,keystone_mdio", "ti,davinci_mdio";
+			#address-cells = <1>;
+			#size-cells = <0>;
+			reg = <0x24200f00 0x100>;
+			status = "disabled";
+			clocks = <&clkcpgmac>;
+			clock-names = "fck";
+			bus_freq	= <2500000>;
+		};
 		/include/ "k2e-netcp.dtsi"
 	};
 };
-
-&mdio {
-	reg = <0x24200f00 0x100>;
-};
--- a/arch/arm/boot/dts/k2hk.dtsi
+++ b/arch/arm/boot/dts/k2hk.dtsi
@ -98,6 +98,17 @@ dspgpio7: keystone_dsp_gpio@262025c {
 			#gpio-cells = <2>;
 			gpio,syscon-dev = <&devctrl 0x25c>;
 		};
+
+		mdio: mdio@02090300 {
+			compatible	= "ti,keystone_mdio", "ti,davinci_mdio";
+			#address-cells = <1>;
+			#size-cells = <0>;
+			reg = <0x02090300 0x100>;
+			status = "disabled";
+			clocks = <&clkcpgmac>;
+			clock-names = "fck";
+			bus_freq	= <2500000>;
+		};
 		/include/ "k2hk-netcp.dtsi"
 	};
 };
--- a/arch/arm/boot/dts/k2l.dtsi
+++ b/arch/arm/boot/dts/k2l.dtsi
@ -29,7 +29,6 @@ cpu@1 {
 	};

 	soc {
-
 		/include/ "k2l-clocks.dtsi"

 		uart2: serial@02348400 {
@ -79,6 +78,17 @@ dspgpio3: keystone_dsp_gpio@262024c {
 			#gpio-cells = <2>;
 			gpio,syscon-dev = <&devctrl 0x24c>;
 		};
+
+		mdio: mdio@26200f00 {
+			compatible	= "ti,keystone_mdio", "ti,davinci_mdio";
+			#address-cells = <1>;
+			#size-cells = <0>;
+			reg = <0x26200f00 0x100>;
+			status = "disabled";
+			clocks = <&clkcpgmac>;
+			clock-names = "fck";
+			bus_freq	= <2500000>;
+		};
 		/include/ "k2l-netcp.dtsi"
 	};
 };
@ -96,7 +106,3 @@ &spi2 {
       /* Pin muxed. Enabled and configured by Bootloader */
       status = "disabled";
 };
-
-&mdio {
-	reg = <0x26200f00 0x100>;
-};
--- a/arch/arm/boot/dts/keystone.dtsi
+++ b/arch/arm/boot/dts/keystone.dtsi
@ -267,17 +267,6 @@ aemif: aemif@21000A00 {
 				  1 0 0x21000A00 0x00000100>;
 		};

-		mdio: mdio@02090300 {
-			compatible	= "ti,keystone_mdio", "ti,davinci_mdio";
-			#address-cells = <1>;
-			#size-cells = <0>;
-			reg		= <0x02090300 0x100>;
-			status = "disabled";
-			clocks = <&clkpa>;
-			clock-names = "fck";
-			bus_freq	= <2500000>;
-		};
-
 		kirq0: keystone_irq@26202a0 {
 			compatible = "ti,keystone-irq";
 			interrupts = <GIC_SPI 4 IRQ_TYPE_EDGE_RISING>;
--- a/arch/arm/boot/dts/omap2430.dtsi
+++ b/arch/arm/boot/dts/omap2430.dtsi
@ -51,7 +51,8 @@ omap2430_pmx: pinmux@30 {
 				};

 				scm_conf: scm_conf@270 {
-					compatible = "syscon";
+					compatible = "syscon",
+						     "simple-bus";
 					reg = <0x270 0x240>;
 					#address-cells = <1>;
 					#size-cells = <1>;
--- a/arch/arm/boot/dts/omap4.dtsi
+++ b/arch/arm/boot/dts/omap4.dtsi
@ -191,7 +191,8 @@ omap4_pmx_core: pinmux@40 {
 				};

 				omap4_padconf_global: omap4_padconf_global@5a0 {
-					compatible = "syscon";
+					compatible = "syscon",
+						     "simple-bus";
 					reg = <0x5a0 0x170>;
 					#address-cells = <1>;
 					#size-cells = <1>;
--- a/arch/arm/boot/dts/omap5.dtsi
+++ b/arch/arm/boot/dts/omap5.dtsi
@ -180,7 +180,8 @@ omap5_pmx_core: pinmux@40 {
 				};

 				omap5_padconf_global: omap5_padconf_global@5a0 {
-					compatible = "syscon";
+					compatible = "syscon",
+						     "simple-bus";
 					reg = <0x5a0 0xec>;
 					#address-cells = <1>;
 					#size-cells = <1>;
--- a/arch/arm/boot/dts/ste-dbx5x0.dtsi
+++ b/arch/arm/boot/dts/ste-dbx5x0.dtsi
@ -15,6 +15,33 @@
 #include "skeleton.dtsi"

 / {
+	cpus {
+		#address-cells = <1>;
+		#size-cells = <0>;
+		enable-method = "ste,dbx500-smp";
+
+		cpu-map {
+			cluster0 {
+				core0 {
+					cpu = <&CPU0>;
+				};
+				core1 {
+					cpu = <&CPU1>;
+				};
+			};
+		};
+		CPU0: cpu@300 {
+			device_type = "cpu";
+			compatible = "arm,cortex-a9";
+			reg = <0x300>;
+		};
+		CPU1: cpu@301 {
+			device_type = "cpu";
+			compatible = "arm,cortex-a9";
+			reg = <0x301>;
+		};
+	};
+
 	soc {
 		#address-cells = <1>;
 		#size-cells = <1>;
@ -22,32 +49,6 @@ soc {
 		interrupt-parent = <&intc>;
 		ranges;

-		cpus {
-			#address-cells = <1>;
-			#size-cells = <0>;
-
-			cpu-map {
-				cluster0 {
-					core0 {
-						cpu = <&CPU0>;
-					};
-					core1 {
-						cpu = <&CPU1>;
-					};
-				};
-			};
-			CPU0: cpu@0 {
-				device_type = "cpu";
-				compatible = "arm,cortex-a9";
-				reg = <0>;
-			};
-			CPU1: cpu@1 {
-				device_type = "cpu";
-				compatible = "arm,cortex-a9";
-				reg = <1>;
-			};
-		};
-
 		ptm@801ae000 {
 			compatible = "arm,coresight-etm3x", "arm,primecell";
 			reg = <0x801ae000 0x1000>;
--- a/arch/arm/configs/exynos_defconfig
+++ b/arch/arm/configs/exynos_defconfig
@ -124,14 +124,14 @@ CONFIG_REGULATOR_S2MPS11=y
 CONFIG_REGULATOR_S5M8767=y
 CONFIG_REGULATOR_TPS65090=y
 CONFIG_DRM=y
-CONFIG_DRM_PTN3460=y
-CONFIG_DRM_PS8622=y
+CONFIG_DRM_NXP_PTN3460=y
+CONFIG_DRM_PARADE_PS8622=y
 CONFIG_DRM_EXYNOS=y
 CONFIG_DRM_EXYNOS_FIMD=y
 CONFIG_DRM_EXYNOS_DSI=y
 CONFIG_DRM_EXYNOS_HDMI=y
 CONFIG_DRM_PANEL_SIMPLE=y
-CONFIG_DRM_PANEL_S6E8AA0=y
+CONFIG_DRM_PANEL_SAMSUNG_S6E8AA0=y
 CONFIG_FB_SIMPLE=y
 CONFIG_EXYNOS_VIDEO=y
 CONFIG_EXYNOS_MIPI_DSI=y
--- a/arch/arm/configs/multi_v7_defconfig
+++ b/arch/arm/configs/multi_v7_defconfig
@ -429,15 +429,15 @@ CONFIG_VIDEO_RENESAS_VSP1=m
 CONFIG_VIDEO_ADV7180=m
 CONFIG_VIDEO_ML86V7667=m
 CONFIG_DRM=y
-CONFIG_DRM_PTN3460=m
-CONFIG_DRM_PS8622=m
+CONFIG_DRM_NXP_PTN3460=m
+CONFIG_DRM_PARADE_PS8622=m
 CONFIG_DRM_EXYNOS=m
 CONFIG_DRM_EXYNOS_DSI=y
 CONFIG_DRM_EXYNOS_FIMD=y
 CONFIG_DRM_EXYNOS_HDMI=y
 CONFIG_DRM_RCAR_DU=m
 CONFIG_DRM_TEGRA=y
-CONFIG_DRM_PANEL_S6E8AA0=m
+CONFIG_DRM_PANEL_SAMSUNG_S6E8AA0=m
 CONFIG_DRM_PANEL_SIMPLE=y
 CONFIG_FB_ARMCLCD=y
 CONFIG_FB_WM8505=y
--- a/arch/arm/kernel/entry-common.S
+++ b/arch/arm/kernel/entry-common.S
@ -61,6 +61,7 @@ work_pending:
 	movlt	scno, #(__NR_restart_syscall - __NR_SYSCALL_BASE)
 	ldmia	sp, {r0 - r6}			@ have to reload r0 - r6
 	b	local_restart			@ ... and off we go
+ENDPROC(ret_fast_syscall)

 /*
 * "slow" syscall return path.  "why" tells us if this was a real syscall.
--- a/arch/arm/kernel/head.S
+++ b/arch/arm/kernel/head.S
@ -399,6 +399,9 @@ ENTRY(secondary_startup)
 	sub	lr, r4, r5			@ mmu has been enabled
 	add	r3, r7, lr
 	ldrd	r4, [r3, #0]			@ get secondary_data.pgdir
+ARM_BE8(eor	r4, r4, r5)			@ Swap r5 and r4 in BE:
+ARM_BE8(eor	r5, r4, r5)			@ it can be done in 3 steps
+ARM_BE8(eor	r4, r4, r5)			@ without using a temp reg.
 	ldr	r8, [r3, #8]			@ get secondary_data.swapper_pg_dir
 	badr	lr, __enable_mmu		@ return address
 	mov	r13, r12			@ __secondary_switched address
--- a/arch/arm/kernel/vdso.c
+++ b/arch/arm/kernel/vdso.c
@ -296,7 +296,6 @@ static bool tk_is_cntvct(const struct timekeeper *tk)
 */
 void update_vsyscall(struct timekeeper *tk)
 {
-	struct timespec xtime_coarse;
 	struct timespec64 *wtm = &tk->wall_to_monotonic;

 	if (!cntvct_ok) {
@ -308,10 +307,10 @@ void update_vsyscall(struct timekeeper *tk)

 	vdso_write_begin(vdso_data);

-	xtime_coarse = __current_kernel_time();
 	vdso_data->tk_is_cntvct			= tk_is_cntvct(tk);
-	vdso_data->xtime_coarse_sec		= xtime_coarse.tv_sec;
-	vdso_data->xtime_coarse_nsec		= xtime_coarse.tv_nsec;
+	vdso_data->xtime_coarse_sec		= tk->xtime_sec;
+	vdso_data->xtime_coarse_nsec		= (u32)(tk->tkr_mono.xtime_nsec >>
+							tk->tkr_mono.shift);
 	vdso_data->wtm_clock_sec		= wtm->tv_sec;
 	vdso_data->wtm_clock_nsec		= wtm->tv_nsec;

--- a/arch/arm/lib/uaccess_with_memcpy.c
+++ b/arch/arm/lib/uaccess_with_memcpy.c
@ -96,7 +96,7 @@ __copy_to_user_memcpy(void __user *to, const void *from, unsigned long n)
 	}

 	/* the mmap semaphore is taken only if not in an atomic context */
-	atomic = in_atomic();
+	atomic = faulthandler_disabled();

 	if (!atomic)
 		down_read(&current->mm->mmap_sem);
--- a/arch/arm/mach-exynos/pm_domains.c
+++ b/arch/arm/mach-exynos/pm_domains.c
@ -146,9 +146,8 @@ static __init int exynos4_pm_init_power_domain(void)
 		pd->base = of_iomap(np, 0);
 		if (!pd->base) {
 			pr_warn("%s: failed to map memory\n", __func__);
-			kfree(pd->pd.name);
+			kfree_const(pd->pd.name);
 			kfree(pd);
-			of_node_put(np);
 			continue;
 		}

--- a/arch/arm/mach-omap2/omap-wakeupgen.c
+++ b/arch/arm/mach-omap2/omap-wakeupgen.c
@ -392,6 +392,7 @@ static struct irq_chip wakeupgen_chip = {
 	.irq_mask		= wakeupgen_mask,
 	.irq_unmask		= wakeupgen_unmask,
 	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_set_type		= irq_chip_set_type_parent,
 	.flags			= IRQCHIP_SKIP_SET_WAKE | IRQCHIP_MASK_ON_SUSPEND,
 #ifdef CONFIG_SMP
 	.irq_set_affinity	= irq_chip_set_affinity_parent,
--- a/arch/arm/vdso/Makefile
+++ b/arch/arm/vdso/Makefile
@ -14,7 +14,7 @@ VDSO_LDFLAGS += -Wl,-z,max-page-size=4096 -Wl,-z,common-page-size=4096
 VDSO_LDFLAGS += -nostdlib -shared
 VDSO_LDFLAGS += $(call cc-ldoption, -Wl$(comma)--hash-style=sysv)
 VDSO_LDFLAGS += $(call cc-ldoption, -Wl$(comma)--build-id)
-VDSO_LDFLAGS += $(call cc-option, -fuse-ld=bfd)
+VDSO_LDFLAGS += $(call cc-ldoption, -fuse-ld=bfd)

 obj-$(CONFIG_VDSO) += vdso.o
 extra-$(CONFIG_VDSO) += vdso.lds
--- a/arch/arm64/kernel/vdso.c
+++ b/arch/arm64/kernel/vdso.c
@ -199,16 +199,15 @@ int arch_setup_additional_pages(struct linux_binprm *bprm,
 */
 void update_vsyscall(struct timekeeper *tk)
 {
-	struct timespec xtime_coarse;
 	u32 use_syscall = strcmp(tk->tkr_mono.clock->name, "arch_sys_counter");

 	++vdso_data->tb_seq_count;
 	smp_wmb();

-	xtime_coarse = __current_kernel_time();
 	vdso_data->use_syscall			= use_syscall;
-	vdso_data->xtime_coarse_sec		= xtime_coarse.tv_sec;
-	vdso_data->xtime_coarse_nsec		= xtime_coarse.tv_nsec;
+	vdso_data->xtime_coarse_sec		= tk->xtime_sec;
+	vdso_data->xtime_coarse_nsec		= tk->tkr_mono.xtime_nsec >>
+							tk->tkr_mono.shift;
 	vdso_data->wtm_clock_sec		= tk->wall_to_monotonic.tv_sec;
 	vdso_data->wtm_clock_nsec		= tk->wall_to_monotonic.tv_nsec;

--- a/arch/mips/kernel/genex.S
+++ b/arch/mips/kernel/genex.S
@ -407,7 +407,7 @@ NESTED(nmi_handler, PT_SIZE, sp)
 	.set	noat
 	SAVE_ALL
 	FEXPORT(handle_\exception\ext)
-	__BUILD_clear_\clear
+	__build_clear_\clear
 	.set	at
 	__BUILD_\verbose \exception
 	move	a0, sp
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@ -80,7 +80,7 @@ syscall_trace_entry:
 	SAVE_STATIC
 	move	s0, t2
 	move	a0, sp
-	daddiu	a1, v0, __NR_64_Linux
+	move	a1, v0
 	jal	syscall_trace_enter

 	bltz	v0, 2f			# seccomp failed? Skip syscall
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@ -72,7 +72,7 @@ n32_syscall_trace_entry:
 	SAVE_STATIC
 	move	s0, t2
 	move	a0, sp
-	daddiu	a1, v0, __NR_N32_Linux
+	move	a1, v0
 	jal	syscall_trace_enter

 	bltz	v0, 2f			# seccomp failed? Skip syscall
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@ -140,6 +140,7 @@ sysexit_from_sys_call:
 	 */
 	andl	$~TS_COMPAT, ASM_THREAD_INFO(TI_status, %rsp, SIZEOF_PTREGS)
 	movl	RIP(%rsp), %ecx		/* User %eip */
+	movq    RAX(%rsp), %rax
 	RESTORE_RSI_RDI
 	xorl	%edx, %edx		/* Do not leak kernel information */
 	xorq	%r8, %r8
@ -219,7 +220,6 @@ sysexit_from_sys_call:
 1:	setbe	%al			/* 1 if error, 0 if not */
 	movzbl	%al, %edi		/* zero-extend that into %edi */
 	call	__audit_syscall_exit
-	movq	RAX(%rsp), %rax		/* reload syscall return value */
 	movl	$(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %edi
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
@ -368,6 +368,7 @@ sysretl_from_sys_call:
 	RESTORE_RSI_RDI_RDX
 	movl	RIP(%rsp), %ecx
 	movl	EFLAGS(%rsp), %r11d
+	movq    RAX(%rsp), %rax
 	xorq	%r10, %r10
 	xorq	%r9, %r9
 	xorq	%r8, %r8
--- a/arch/x86/include/asm/sigcontext.h
+++ b/arch/x86/include/asm/sigcontext.h
@ -57,9 +57,9 @@ struct sigcontext {
 	unsigned long ip;
 	unsigned long flags;
 	unsigned short cs;
-	unsigned short __pad2;	/* Was called gs, but was always zero. */
-	unsigned short __pad1;	/* Was called fs, but was always zero. */
-	unsigned short ss;
+	unsigned short gs;
+	unsigned short fs;
+	unsigned short __pad0;
 	unsigned long err;
 	unsigned long trapno;
 	unsigned long oldmask;
--- a/arch/x86/include/asm/switch_to.h
+++ b/arch/x86/include/asm/switch_to.h
@ -79,12 +79,12 @@ do {									\
 #else /* CONFIG_X86_32 */

 /* frame pointer must be last for get_wchan */
-#define SAVE_CONTEXT    "pushq %%rbp ; movq %%rsi,%%rbp\n\t"
-#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp\t"
+#define SAVE_CONTEXT    "pushf ; pushq %%rbp ; movq %%rsi,%%rbp\n\t"
+#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp ; popf\t"

 #define __EXTRA_CLOBBER  \
 	, "rcx", "rbx", "rdx", "r8", "r9", "r10", "r11", \
-	  "r12", "r13", "r14", "r15", "flags"
+	  "r12", "r13", "r14", "r15"

 #ifdef CONFIG_CC_STACKPROTECTOR
 #define __switch_canary							  \
@ -100,11 +100,7 @@ do {									\
 #define __switch_canary_iparam
 #endif	/* CC_STACKPROTECTOR */

-/*
- * There is no need to save or restore flags, because flags are always
- * clean in kernel mode, with the possible exception of IOPL.  Kernel IOPL
- * has no effect.
- */
+/* Save restore flags to clear handle leaking NT */
 #define switch_to(prev, next, last) \
 	asm volatile(SAVE_CONTEXT					  \
 	     "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */	  \
--- a/arch/x86/include/uapi/asm/sigcontext.h
+++ b/arch/x86/include/uapi/asm/sigcontext.h
@ -177,24 +177,9 @@ struct sigcontext {
 	__u64 rip;
 	__u64 eflags;		/* RFLAGS */
 	__u16 cs;
-
-	/*
-	 * Prior to 2.5.64 ("[PATCH] x86-64 updates for 2.5.64-bk3"),
-	 * Linux saved and restored fs and gs in these slots.  This
-	 * was counterproductive, as fsbase and gsbase were never
-	 * saved, so arch_prctl was presumably unreliable.
-	 *
-	 * If these slots are ever needed for any other purpose, there
-	 * is some risk that very old 64-bit binaries could get
-	 * confused.  I doubt that many such binaries still work,
-	 * though, since the same patch in 2.5.64 also removed the
-	 * 64-bit set_thread_area syscall, so it appears that there is
-	 * no TLS API that works in both pre- and post-2.5.64 kernels.
-	 */
-	__u16 __pad2;		/* Was gs. */
-	__u16 __pad1;		/* Was fs. */
-
-	__u16 ss;
+	__u16 gs;
+	__u16 fs;
+	__u16 __pad0;
 	__u64 err;
 	__u64 trapno;
 	__u64 oldmask;
--- a/arch/x86/kernel/apic/vector.c
+++ b/arch/x86/kernel/apic/vector.c
@ -322,7 +322,7 @@ static int x86_vector_alloc_irqs(struct irq_domain *domain, unsigned int virq,
 		irq_data->chip = &lapic_controller;
 		irq_data->chip_data = data;
 		irq_data->hwirq = virq + i;
-		err = assign_irq_vector_policy(virq, irq_data->node, data,
+		err = assign_irq_vector_policy(virq + i, irq_data->node, data,
 					       info);
 		if (err)
 			goto error;
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@ -2534,7 +2534,7 @@ static int intel_pmu_cpu_prepare(int cpu)
 	if (x86_pmu.extra_regs || x86_pmu.lbr_sel_map) {
 		cpuc->shared_regs = allocate_shared_regs(cpu);
 		if (!cpuc->shared_regs)
-			return NOTIFY_BAD;
+			goto err;
 	}

 	if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) {
@ -2542,18 +2542,27 @@ static int intel_pmu_cpu_prepare(int cpu)

 		cpuc->constraint_list = kzalloc(sz, GFP_KERNEL);
 		if (!cpuc->constraint_list)
-			return NOTIFY_BAD;
+			goto err_shared_regs;

 		cpuc->excl_cntrs = allocate_excl_cntrs(cpu);
-		if (!cpuc->excl_cntrs) {
-			kfree(cpuc->constraint_list);
-			kfree(cpuc->shared_regs);
-			return NOTIFY_BAD;
-		}
+		if (!cpuc->excl_cntrs)
+			goto err_constraint_list;
+
 		cpuc->excl_thread_id = 0;
 	}

 	return NOTIFY_OK;
+
+err_constraint_list:
+	kfree(cpuc->constraint_list);
+	cpuc->constraint_list = NULL;
+
+err_shared_regs:
+	kfree(cpuc->shared_regs);
+	cpuc->shared_regs = NULL;
+
+err:
+	return NOTIFY_BAD;
 }

 static void intel_pmu_cpu_starting(int cpu)
--- a/arch/x86/kernel/cpu/perf_event_intel_cqm.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_cqm.c
@ -1255,7 +1255,7 @@ static inline void cqm_pick_event_reader(int cpu)
 	cpumask_set_cpu(cpu, &cqm_cpumask);
 }

-static void intel_cqm_cpu_prepare(unsigned int cpu)
+static void intel_cqm_cpu_starting(unsigned int cpu)
 {
 	struct intel_pqr_state *state = &per_cpu(pqr_state, cpu);
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
@ -1296,13 +1296,11 @@ static int intel_cqm_cpu_notifier(struct notifier_block *nb,
 	unsigned int cpu  = (unsigned long)hcpu;

 	switch (action & ~CPU_TASKS_FROZEN) {
-	case CPU_UP_PREPARE:
-		intel_cqm_cpu_prepare(cpu);
-		break;
 	case CPU_DOWN_PREPARE:
 		intel_cqm_cpu_exit(cpu);
 		break;
 	case CPU_STARTING:
+		intel_cqm_cpu_starting(cpu);
 		cqm_pick_event_reader(cpu);
 		break;
 	}
@ -1373,7 +1371,7 @@ static int __init intel_cqm_init(void)
 		goto out;

 	for_each_online_cpu(i) {
-		intel_cqm_cpu_prepare(i);
+		intel_cqm_cpu_starting(i);
 		cqm_pick_event_reader(i);
 	}

--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@ -270,7 +270,7 @@ int fpu__copy(struct fpu *dst_fpu, struct fpu *src_fpu)
 	dst_fpu->fpregs_active = 0;
 	dst_fpu->last_cpu = -1;

-	if (src_fpu->fpstate_active)
+	if (src_fpu->fpstate_active && cpu_has_fpu)
 		fpu_copy(dst_fpu, src_fpu);

 	return 0;
--- a/arch/x86/kernel/fpu/init.c
+++ b/arch/x86/kernel/fpu/init.c
@ -40,7 +40,12 @@ static void fpu__init_cpu_generic(void)
 	write_cr0(cr0);

 	/* Flush out any pending x87 state: */
-	asm volatile ("fninit");
+#ifdef CONFIG_MATH_EMULATION
+	if (!cpu_has_fpu)
+		fpstate_init_soft(&current->thread.fpu.state.soft);
+	else
+#endif
+		asm volatile ("fninit");
 }

 /*
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@ -408,6 +408,7 @@ static int prefer_mwait_c1_over_halt(const struct cpuinfo_x86 *c)
 static void mwait_idle(void)
 {
 	if (!current_set_polling_and_test()) {
+		trace_cpu_idle_rcuidle(1, smp_processor_id());
 		if (this_cpu_has(X86_BUG_CLFLUSH_MONITOR)) {
 			smp_mb(); /* quirk */
 			clflush((void *)&current_thread_info()->flags);
@ -419,6 +420,7 @@ static void mwait_idle(void)
 			__sti_mwait(0, 0);
 		else
 			local_irq_enable();
+		trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, smp_processor_id());
 	} else {
 		local_irq_enable();
 	}
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@ -93,8 +93,15 @@ int restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc)
 		COPY(r15);
 #endif /* CONFIG_X86_64 */

+#ifdef CONFIG_X86_32
 		COPY_SEG_CPL3(cs);
 		COPY_SEG_CPL3(ss);
+#else /* !CONFIG_X86_32 */
+		/* Kernel saves and restores only the CS segment register on signals,
+		 * which is the bare minimum needed to allow mixed 32/64-bit code.
+		 * App's signal handler can save/restore other segments if needed. */
+		COPY_SEG_CPL3(cs);
+#endif /* CONFIG_X86_32 */

 		get_user_ex(tmpflags, &sc->flags);
 		regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
@ -154,9 +161,8 @@ int setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate,
 #else /* !CONFIG_X86_32 */
 		put_user_ex(regs->flags, &sc->flags);
 		put_user_ex(regs->cs, &sc->cs);
-		put_user_ex(0, &sc->__pad2);
-		put_user_ex(0, &sc->__pad1);
-		put_user_ex(regs->ss, &sc->ss);
+		put_user_ex(0, &sc->gs);
+		put_user_ex(0, &sc->fs);
 #endif /* CONFIG_X86_32 */

 		put_user_ex(fpstate, &sc->fpstate);
@ -451,19 +457,9 @@ static int __setup_rt_frame(int sig, struct ksignal *ksig,

 	regs->sp = (unsigned long)frame;

-	/*
-	 * Set up the CS and SS registers to run signal handlers in
-	 * 64-bit mode, even if the handler happens to be interrupting
-	 * 32-bit or 16-bit code.
-	 *
-	 * SS is subtle.  In 64-bit mode, we don't need any particular
-	 * SS descriptor, but we do need SS to be valid.  It's possible
-	 * that the old SS is entirely bogus -- this can happen if the
-	 * signal we're trying to deliver is #GP or #SS caused by a bad
-	 * SS value.
-	 */
+	/* Set up the CS register to run signal handlers in 64-bit mode,
+	   even if the handler happens to be interrupting 32-bit code. */
 	regs->cs = __USER_CS;
-	regs->ss = __USER_DS;

 	return 0;
 }
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@ -28,11 +28,11 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re
 		struct desc_struct *desc;
 		unsigned long base;

-		seg &= ~7UL;
+		seg >>= 3;

 		mutex_lock(&child->mm->context.lock);
 		if (unlikely(!child->mm->context.ldt ||
-			     (seg >> 3) >= child->mm->context.ldt->size))
+			     seg >= child->mm->context.ldt->size))
 			addr = -1L; /* bogus selector, access would fault */
 		else {
 			desc = &child->mm->context.ldt->entries[seg];
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@ -2105,7 +2105,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		if (guest_cpuid_has_tsc_adjust(vcpu)) {
 			if (!msr_info->host_initiated) {
 				s64 adj = data - vcpu->arch.ia32_tsc_adjust_msr;
-				kvm_x86_ops->adjust_tsc_offset(vcpu, adj, true);
+				adjust_tsc_offset_guest(vcpu, adj);
 			}
 			vcpu->arch.ia32_tsc_adjust_msr = data;
 		}
@ -6327,6 +6327,7 @@ static void process_smi_save_state_64(struct kvm_vcpu *vcpu, char *buf)
 static void process_smi(struct kvm_vcpu *vcpu)
 {
 	struct kvm_segment cs, ds;
+	struct desc_ptr dt;
 	char buf[512];
 	u32 cr0;

@ -6359,6 +6360,10 @@ static void process_smi(struct kvm_vcpu *vcpu)

 	kvm_x86_ops->set_cr4(vcpu, 0);

+	/* Undocumented: IDT limit is set to zero on entry to SMM.  */
+	dt.address = dt.size = 0;
+	kvm_x86_ops->set_idt(vcpu, &dt);
+
 	__kvm_set_dr(vcpu, 7, DR7_FIXED_1);

 	cs.selector = (vcpu->arch.smbase >> 4) & 0xffff;
--- a/arch/x86/math-emu/fpu_entry.c
+++ b/arch/x86/math-emu/fpu_entry.c
@ -29,7 +29,6 @@

 #include <asm/uaccess.h>
 #include <asm/traps.h>
-#include <asm/desc.h>
 #include <asm/user.h>
 #include <asm/fpu/internal.h>

@ -181,7 +180,7 @@ void math_emulate(struct math_emu_info *info)
 			math_abort(FPU_info, SIGILL);
 		}

-		code_descriptor = LDT_DESCRIPTOR(FPU_CS);
+		code_descriptor = FPU_get_ldt_descriptor(FPU_CS);
 		if (SEG_D_SIZE(code_descriptor)) {
 			/* The above test may be wrong, the book is not clear */
 			/* Segmented 32 bit protected mode */
--- a/arch/x86/math-emu/fpu_system.h
+++ b/arch/x86/math-emu/fpu_system.h
@ -16,9 +16,24 @@
 #include <linux/kernel.h>
 #include <linux/mm.h>

-/* s is always from a cpu register, and the cpu does bounds checking
- * during register load --> no further bounds checks needed */
-#define LDT_DESCRIPTOR(s)	(((struct desc_struct *)current->mm->context.ldt)[(s) >> 3])
+#include <asm/desc.h>
+#include <asm/mmu_context.h>
+
+static inline struct desc_struct FPU_get_ldt_descriptor(unsigned seg)
+{
+	static struct desc_struct zero_desc;
+	struct desc_struct ret = zero_desc;
+
+#ifdef CONFIG_MODIFY_LDT_SYSCALL
+	seg >>= 3;
+	mutex_lock(&current->mm->context.lock);
+	if (current->mm->context.ldt && seg < current->mm->context.ldt->size)
+		ret = current->mm->context.ldt->entries[seg];
+	mutex_unlock(&current->mm->context.lock);
+#endif
+	return ret;
+}
+
 #define SEG_D_SIZE(x)		((x).b & (3 << 21))
 #define SEG_G_BIT(x)		((x).b & (1 << 23))
 #define SEG_GRANULARITY(x)	(((x).b & (1 << 23)) ? 4096 : 1)
--- a/arch/x86/math-emu/get_address.c
+++ b/arch/x86/math-emu/get_address.c
@ -20,7 +20,6 @@
 #include <linux/stddef.h>

 #include <asm/uaccess.h>
-#include <asm/desc.h>

 #include "fpu_system.h"
 #include "exception.h"
@ -158,7 +157,7 @@ static long pm_address(u_char FPU_modrm, u_char segment,
 		addr->selector = PM_REG_(segment);
 	}

-	descriptor = LDT_DESCRIPTOR(PM_REG_(segment));
+	descriptor = FPU_get_ldt_descriptor(addr->selector);
 	base_address = SEG_BASE_ADDR(descriptor);
 	address = base_address + offset;
 	limit = base_address
--- a/arch/x86/xen/Kconfig
+++ b/arch/x86/xen/Kconfig
@ -8,7 +8,7 @@ config XEN
 	select PARAVIRT_CLOCK
 	select XEN_HAVE_PVMMU
 	depends on X86_64 || (X86_32 && X86_PAE)
-	depends on X86_TSC
+	depends on X86_LOCAL_APIC && X86_TSC
 	help
 	  This is the Linux Xen port.  Enabling this will allow the
 	  kernel to boot in a paravirtualized environment under the
@ -17,7 +17,7 @@ config XEN
 config XEN_DOM0
 	def_bool y
 	depends on XEN && PCI_XEN && SWIOTLB_XEN
-	depends on X86_LOCAL_APIC && X86_IO_APIC && ACPI && PCI
+	depends on X86_IO_APIC && ACPI && PCI

 config XEN_PVHVM
 	def_bool y
--- a/arch/x86/xen/Makefile
+++ b/arch/x86/xen/Makefile
@ -13,13 +13,13 @@ CFLAGS_mmu.o			:= $(nostackp)
 obj-y		:= enlighten.o setup.o multicalls.o mmu.o irq.o \
 			time.o xen-asm.o xen-asm_$(BITS).o \
 			grant-table.o suspend.o platform-pci-unplug.o \
-			p2m.o
+			p2m.o apic.o

 obj-$(CONFIG_EVENT_TRACING) += trace.o

 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= spinlock.o
 obj-$(CONFIG_XEN_DEBUG_FS)	+= debugfs.o
-obj-$(CONFIG_XEN_DOM0)		+= apic.o vga.o
+obj-$(CONFIG_XEN_DOM0)		+= vga.o
 obj-$(CONFIG_SWIOTLB_XEN)	+= pci-swiotlb-xen.o
 obj-$(CONFIG_XEN_EFI)		+= efi.o
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@ -101,17 +101,15 @@ struct dom0_vga_console_info;

 #ifdef CONFIG_XEN_DOM0
 void __init xen_init_vga(const struct dom0_vga_console_info *, size_t size);
-void __init xen_init_apic(void);
 #else
 static inline void __init xen_init_vga(const struct dom0_vga_console_info *info,
 				       size_t size)
 {
 }
-static inline void __init xen_init_apic(void)
-{
-}
 #endif

+void __init xen_init_apic(void);
+
 #ifdef CONFIG_XEN_EFI
 extern void xen_efi_init(void);
 #else
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@ -241,8 +241,8 @@ EXPORT_SYMBOL(blk_queue_bounce_limit);
 * Description:
 *    Enables a low level driver to set a hard upper limit,
 *    max_hw_sectors, on the size of requests.  max_hw_sectors is set by
- *    the device driver based upon the combined capabilities of I/O
- *    controller and storage device.
+ *    the device driver based upon the capabilities of the I/O
+ *    controller.
 *
 *    max_sectors is a soft limit imposed by the block layer for
 *    filesystem type requests.  This value can be overridden on a
--- a/crypto/authencesn.c
+++ b/crypto/authencesn.c
@ -393,8 +393,6 @@ static int crypto_authenc_esn_genicv(struct aead_request *req, u8 *iv,
 	struct scatterlist *cipher = areq_ctx->cipher;
 	struct scatterlist *hsg = areq_ctx->hsg;
 	struct scatterlist *tsg = areq_ctx->tsg;
-	struct scatterlist *assoc1;
-	struct scatterlist *assoc2;
 	unsigned int ivsize = crypto_aead_ivsize(authenc_esn);
 	unsigned int cryptlen = req->cryptlen;
 	struct page *dstp;
@ -412,27 +410,19 @@ static int crypto_authenc_esn_genicv(struct aead_request *req, u8 *iv,
 		cryptlen += ivsize;
 	}

-	if (sg_is_last(assoc))
-		return -EINVAL;
-
-	assoc1 = assoc + 1;
-	if (sg_is_last(assoc1))
-		return -EINVAL;
-
-	assoc2 = assoc + 2;
-	if (!sg_is_last(assoc2))
+	if (assoc->length < 12)
 		return -EINVAL;

 	sg_init_table(hsg, 2);
-	sg_set_page(hsg, sg_page(assoc), assoc->length, assoc->offset);
-	sg_set_page(hsg + 1, sg_page(assoc2), assoc2->length, assoc2->offset);
+	sg_set_page(hsg, sg_page(assoc), 4, assoc->offset);
+	sg_set_page(hsg + 1, sg_page(assoc), 4, assoc->offset + 8);

 	sg_init_table(tsg, 1);
-	sg_set_page(tsg, sg_page(assoc1), assoc1->length, assoc1->offset);
+	sg_set_page(tsg, sg_page(assoc), 4, assoc->offset + 4);

 	areq_ctx->cryptlen = cryptlen;
-	areq_ctx->headlen = assoc->length + assoc2->length;
-	areq_ctx->trailen = assoc1->length;
+	areq_ctx->headlen = 8;
+	areq_ctx->trailen = 4;
 	areq_ctx->sg = dst;

 	areq_ctx->complete = authenc_esn_geniv_ahash_done;
@ -563,8 +553,6 @@ static int crypto_authenc_esn_iverify(struct aead_request *req, u8 *iv,
 	struct scatterlist *cipher = areq_ctx->cipher;
 	struct scatterlist *hsg = areq_ctx->hsg;
 	struct scatterlist *tsg = areq_ctx->tsg;
-	struct scatterlist *assoc1;
-	struct scatterlist *assoc2;
 	unsigned int ivsize = crypto_aead_ivsize(authenc_esn);
 	struct page *srcp;
 	u8 *vsrc;
@ -580,27 +568,19 @@ static int crypto_authenc_esn_iverify(struct aead_request *req, u8 *iv,
 		cryptlen += ivsize;
 	}

-	if (sg_is_last(assoc))
-		return -EINVAL;
-
-	assoc1 = assoc + 1;
-	if (sg_is_last(assoc1))
-		return -EINVAL;
-
-	assoc2 = assoc + 2;
-	if (!sg_is_last(assoc2))
+	if (assoc->length < 12)
 		return -EINVAL;

 	sg_init_table(hsg, 2);
-	sg_set_page(hsg, sg_page(assoc), assoc->length, assoc->offset);
-	sg_set_page(hsg + 1, sg_page(assoc2), assoc2->length, assoc2->offset);
+	sg_set_page(hsg, sg_page(assoc), 4, assoc->offset);
+	sg_set_page(hsg + 1, sg_page(assoc), 4, assoc->offset + 8);

 	sg_init_table(tsg, 1);
-	sg_set_page(tsg, sg_page(assoc1), assoc1->length, assoc1->offset);
+	sg_set_page(tsg, sg_page(assoc), 4, assoc->offset + 4);

 	areq_ctx->cryptlen = cryptlen;
-	areq_ctx->headlen = assoc->length + assoc2->length;
-	areq_ctx->trailen = assoc1->length;
+	areq_ctx->headlen = 8;
+	areq_ctx->trailen = 4;
 	areq_ctx->sg = src;

 	areq_ctx->complete = authenc_esn_verify_ahash_done;
--- a/drivers/acpi/video_detect.c
+++ b/drivers/acpi/video_detect.c
@ -32,6 +32,7 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/types.h>
+#include <linux/workqueue.h>
 #include <acpi/video.h>

 ACPI_MODULE_NAME("video");
@ -41,6 +42,7 @@ void acpi_video_unregister_backlight(void);

 static bool backlight_notifier_registered;
 static struct notifier_block backlight_nb;
+static struct work_struct backlight_notify_work;

 static enum acpi_backlight_type acpi_backlight_cmdline = acpi_backlight_undef;
 static enum acpi_backlight_type acpi_backlight_dmi = acpi_backlight_undef;
@ -262,6 +264,13 @@ static const struct dmi_system_id video_detect_dmi_table[] = {
 	{ },
 };

+/* This uses a workqueue to avoid various locking ordering issues */
+static void acpi_video_backlight_notify_work(struct work_struct *work)
+{
+	if (acpi_video_get_backlight_type() != acpi_backlight_video)
+		acpi_video_unregister_backlight();
+}
+
 static int acpi_video_backlight_notify(struct notifier_block *nb,
 				       unsigned long val, void *bd)
 {
@ -269,9 +278,8 @@ static int acpi_video_backlight_notify(struct notifier_block *nb,

 	/* A raw bl registering may change video -> native */
 	if (backlight->props.type == BACKLIGHT_RAW &&
-	    val == BACKLIGHT_REGISTERED &&
-	    acpi_video_get_backlight_type() != acpi_backlight_video)
-		acpi_video_unregister_backlight();
+	    val == BACKLIGHT_REGISTERED)
+		schedule_work(&backlight_notify_work);

 	return NOTIFY_OK;
 }
@ -304,6 +312,8 @@ enum acpi_backlight_type acpi_video_get_backlight_type(void)
 		acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
 				    ACPI_UINT32_MAX, find_video, NULL,
 				    &video_caps, NULL);
+		INIT_WORK(&backlight_notify_work,
+			  acpi_video_backlight_notify_work);
 		backlight_nb.notifier_call = acpi_video_backlight_notify;
 		backlight_nb.priority = 0;
 		if (backlight_register_notifier(&backlight_nb) == 0)
--- a/drivers/ata/ahci_brcmstb.c
+++ b/drivers/ata/ahci_brcmstb.c
@ -92,7 +92,7 @@ static inline u32 brcm_sata_readreg(void __iomem *addr)
 	 * Other architectures (e.g., ARM) either do not support big endian, or
 	 * else leave I/O in little endian mode.
 	 */
-	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(__BIG_ENDIAN))
+	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
 		return __raw_readl(addr);
 	else
 		return readl_relaxed(addr);
@ -101,7 +101,7 @@ static inline u32 brcm_sata_readreg(void __iomem *addr)
 static inline void brcm_sata_writereg(u32 val, void __iomem *addr)
 {
 	/* See brcm_sata_readreg() comments */
-	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(__BIG_ENDIAN))
+	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
 		__raw_writel(val, addr);
 	else
 		writel_relaxed(val, addr);
@ -209,6 +209,7 @@ static void brcm_sata_init(struct brcm_ahci_priv *priv)
 			   priv->top_ctrl + SATA_TOP_CTRL_BUS_CTRL);
 }

+#ifdef CONFIG_PM_SLEEP
 static int brcm_ahci_suspend(struct device *dev)
 {
 	struct ata_host *host = dev_get_drvdata(dev);
@ -231,6 +232,7 @@ static int brcm_ahci_resume(struct device *dev)
 	brcm_sata_phys_enable(priv);
 	return ahci_platform_resume(dev);
 }
+#endif

 static struct scsi_host_template ahci_platform_sht = {
 	AHCI_SHT(DRV_NAME),
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@ -694,11 +694,11 @@ static int ata_rwcmd_protocol(struct ata_taskfile *tf, struct ata_device *dev)
 *	RETURNS:
 *	Block address read from @tf.
 */
-u64 ata_tf_read_block(const struct ata_taskfile *tf, struct ata_device *dev)
+u64 ata_tf_read_block(struct ata_taskfile *tf, struct ata_device *dev)
 {
 	u64 block = 0;

-	if (!dev || tf->flags & ATA_TFLAG_LBA) {
+	if (tf->flags & ATA_TFLAG_LBA) {
 		if (tf->flags & ATA_TFLAG_LBA48) {
 			block |= (u64)tf->hob_lbah << 40;
 			block |= (u64)tf->hob_lbam << 32;
@ -2147,24 +2147,6 @@ static int ata_dev_config_ncq(struct ata_device *dev,
 	return 0;
 }

-static void ata_dev_config_sense_reporting(struct ata_device *dev)
-{
-	unsigned int err_mask;
-
-	if (!ata_id_has_sense_reporting(dev->id))
-		return;
-
-	if (ata_id_sense_reporting_enabled(dev->id))
-		return;
-
-	err_mask = ata_dev_set_feature(dev, SETFEATURE_SENSE_DATA, 0x1);
-	if (err_mask) {
-		ata_dev_dbg(dev,
-			    "failed to enable Sense Data Reporting, Emask 0x%x\n",
-			    err_mask);
-	}
-}
-
 /**
 *	ata_dev_configure - Configure the specified ATA/ATAPI device
 *	@dev: Target device to configure
@ -2387,7 +2369,7 @@ int ata_dev_configure(struct ata_device *dev)
 					dev->devslp_timing[i] = sata_setting[j];
 				}
 		}
-		ata_dev_config_sense_reporting(dev);
+
 		dev->cdb_len = 16;
 	}

--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@ -1592,8 +1592,6 @@ static int ata_eh_read_log_10h(struct ata_device *dev,
 	tf->hob_lbah = buf[10];
 	tf->nsect = buf[12];
 	tf->hob_nsect = buf[13];
-	if (ata_id_has_ncq_autosense(dev->id))
-		tf->auxiliary = buf[14] << 16 | buf[15] << 8 | buf[16];

 	return 0;
 }
@ -1629,70 +1627,6 @@ unsigned int atapi_eh_tur(struct ata_device *dev, u8 *r_sense_key)
 	return err_mask;
 }

-/**
- *	ata_eh_request_sense - perform REQUEST_SENSE_DATA_EXT
- *	@dev: device to perform REQUEST_SENSE_SENSE_DATA_EXT to
- *	@sense_buf: result sense data buffer (SCSI_SENSE_BUFFERSIZE bytes long)
- *	@dfl_sense_key: default sense key to use
- *
- *	Perform REQUEST_SENSE_DATA_EXT after the device reported CHECK
- *	SENSE.  This function is EH helper.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep).
- *
- *	RETURNS:
- *	encoded sense data on success, 0 on failure or if sense data
- *	is not available.
- */
-static u32 ata_eh_request_sense(struct ata_queued_cmd *qc,
-				struct scsi_cmnd *cmd)
-{
-	struct ata_device *dev = qc->dev;
-	struct ata_taskfile tf;
-	unsigned int err_mask;
-
-	if (!cmd)
-		return 0;
-
-	DPRINTK("ATA request sense\n");
-	ata_dev_warn(dev, "request sense\n");
-	if (!ata_id_sense_reporting_enabled(dev->id)) {
-		ata_dev_warn(qc->dev, "sense data reporting disabled\n");
-		return 0;
-	}
-	ata_tf_init(dev, &tf);
-
-	tf.flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE;
-	tf.flags |= ATA_TFLAG_LBA | ATA_TFLAG_LBA48;
-	tf.command = ATA_CMD_REQ_SENSE_DATA;
-	tf.protocol = ATA_PROT_NODATA;
-
-	err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 0);
-	/*
-	 * ACS-4 states:
-	 * The device may set the SENSE DATA AVAILABLE bit to one in the
-	 * STATUS field and clear the ERROR bit to zero in the STATUS field
-	 * to indicate that the command returned completion without an error
-	 * and the sense data described in table 306 is available.
-	 *
-	 * IOW the 'ATA_SENSE' bit might not be set even though valid
-	 * sense data is available.
-	 * So check for both.
-	 */
-	if ((tf.command & ATA_SENSE) ||
-		tf.lbah != 0 || tf.lbam != 0 || tf.lbal != 0) {
-		ata_scsi_set_sense(cmd, tf.lbah, tf.lbam, tf.lbal);
-		qc->flags |= ATA_QCFLAG_SENSE_VALID;
-		ata_dev_warn(dev, "sense data %02x/%02x/%02x\n",
-			     tf.lbah, tf.lbam, tf.lbal);
-	} else {
-		ata_dev_warn(dev, "request sense failed stat %02x emask %x\n",
-			     tf.command, err_mask);
-	}
-	return err_mask;
-}
-
 /**
 *	atapi_eh_request_sense - perform ATAPI REQUEST_SENSE
 *	@dev: device to perform REQUEST_SENSE to
@ -1855,19 +1789,6 @@ void ata_eh_analyze_ncq_error(struct ata_link *link)
 	memcpy(&qc->result_tf, &tf, sizeof(tf));
 	qc->result_tf.flags = ATA_TFLAG_ISADDR | ATA_TFLAG_LBA | ATA_TFLAG_LBA48;
 	qc->err_mask |= AC_ERR_DEV | AC_ERR_NCQ;
-	if (qc->result_tf.auxiliary) {
-		char sense_key, asc, ascq;
-
-		sense_key = (qc->result_tf.auxiliary >> 16) & 0xff;
-		asc = (qc->result_tf.auxiliary >> 8) & 0xff;
-		ascq = qc->result_tf.auxiliary & 0xff;
-		ata_dev_dbg(dev, "NCQ Autosense %02x/%02x/%02x\n",
-			    sense_key, asc, ascq);
-		ata_scsi_set_sense(qc->scsicmd, sense_key, asc, ascq);
-		ata_scsi_set_sense_information(qc->scsicmd, &qc->result_tf);
-		qc->flags |= ATA_QCFLAG_SENSE_VALID;
-	}
-
 	ehc->i.err_mask &= ~AC_ERR_DEV;
 }

@ -1897,27 +1818,6 @@ static unsigned int ata_eh_analyze_tf(struct ata_queued_cmd *qc,
 		return ATA_EH_RESET;
 	}

-	/*
-	 * Sense data reporting does not work if the
-	 * device fault bit is set.
-	 */
-	if ((stat & ATA_SENSE) && !(stat & ATA_DF) &&
-	    !(qc->flags & ATA_QCFLAG_SENSE_VALID)) {
-		if (!(qc->ap->pflags & ATA_PFLAG_FROZEN)) {
-			tmp = ata_eh_request_sense(qc, qc->scsicmd);
-			if (tmp)
-				qc->err_mask |= tmp;
-			else
-				ata_scsi_set_sense_information(qc->scsicmd, tf);
-		} else {
-			ata_dev_warn(qc->dev, "sense data available but port frozen\n");
-		}
-	}
-
-	/* Set by NCQ autosense or request sense above */
-	if (qc->flags & ATA_QCFLAG_SENSE_VALID)
-		return 0;
-
 	if (stat & (ATA_ERR | ATA_DF))
 		qc->err_mask |= AC_ERR_DEV;
 	else
@ -2661,15 +2561,14 @@ static void ata_eh_link_report(struct ata_link *link)

 #ifdef CONFIG_ATA_VERBOSE_ERROR
 		if (res->command & (ATA_BUSY | ATA_DRDY | ATA_DF | ATA_DRQ |
-				    ATA_SENSE | ATA_ERR)) {
+				    ATA_ERR)) {
 			if (res->command & ATA_BUSY)
 				ata_dev_err(qc->dev, "status: { Busy }\n");
 			else
-				ata_dev_err(qc->dev, "status: { %s%s%s%s%s}\n",
+				ata_dev_err(qc->dev, "status: { %s%s%s%s}\n",
 				  res->command & ATA_DRDY ? "DRDY " : "",
 				  res->command & ATA_DF ? "DF " : "",
 				  res->command & ATA_DRQ ? "DRQ " : "",
-				  res->command & ATA_SENSE ? "SENSE " : "",
 				  res->command & ATA_ERR ? "ERR " : "");
 		}

--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@ -270,28 +270,13 @@ DEVICE_ATTR(unload_heads, S_IRUGO | S_IWUSR,
 	    ata_scsi_park_show, ata_scsi_park_store);
 EXPORT_SYMBOL_GPL(dev_attr_unload_heads);

-void ata_scsi_set_sense(struct scsi_cmnd *cmd, u8 sk, u8 asc, u8 ascq)
+static void ata_scsi_set_sense(struct scsi_cmnd *cmd, u8 sk, u8 asc, u8 ascq)
 {
-	if (!cmd)
-		return;
-
 	cmd->result = (DRIVER_SENSE << 24) | SAM_STAT_CHECK_CONDITION;

 	scsi_build_sense_buffer(0, cmd->sense_buffer, sk, asc, ascq);
 }

-void ata_scsi_set_sense_information(struct scsi_cmnd *cmd,
-				    const struct ata_taskfile *tf)
-{
-	u64 information;
-
-	if (!cmd)
-		return;
-
-	information = ata_tf_read_block(tf, NULL);
-	scsi_set_sense_information(cmd->sense_buffer, information);
-}
-
 static ssize_t
 ata_scsi_em_message_store(struct device *dev, struct device_attribute *attr,
 			  const char *buf, size_t count)
@ -1792,9 +1777,7 @@ static void ata_scsi_qc_complete(struct ata_queued_cmd *qc)
 	    ((cdb[2] & 0x20) || need_sense)) {
 		ata_gen_passthru_sense(qc);
 	} else {
-		if (qc->flags & ATA_QCFLAG_SENSE_VALID) {
-			cmd->result = SAM_STAT_CHECK_CONDITION;
-		} else if (!need_sense) {
+		if (!need_sense) {
 			cmd->result = SAM_STAT_GOOD;
 		} else {
 			/* TODO: decide which descriptor format to use
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@ -67,8 +67,7 @@ extern struct ata_queued_cmd *ata_qc_new_init(struct ata_device *dev, int tag);
 extern int ata_build_rw_tf(struct ata_taskfile *tf, struct ata_device *dev,
 			   u64 block, u32 n_block, unsigned int tf_flags,
 			   unsigned int tag);
-extern u64 ata_tf_read_block(const struct ata_taskfile *tf,
-			     struct ata_device *dev);
+extern u64 ata_tf_read_block(struct ata_taskfile *tf, struct ata_device *dev);
 extern unsigned ata_exec_internal(struct ata_device *dev,
 				  struct ata_taskfile *tf, const u8 *cdb,
 				  int dma_dir, void *buf, unsigned int buflen,
@ -138,9 +137,6 @@ extern int ata_scsi_add_hosts(struct ata_host *host,
 			      struct scsi_host_template *sht);
 extern void ata_scsi_scan_host(struct ata_port *ap, int sync);
 extern int ata_scsi_offline_dev(struct ata_device *dev);
-extern void ata_scsi_set_sense(struct scsi_cmnd *cmd, u8 sk, u8 asc, u8 ascq);
-extern void ata_scsi_set_sense_information(struct scsi_cmnd *cmd,
-					   const struct ata_taskfile *tf);
 extern void ata_scsi_media_change_notify(struct ata_device *dev);
 extern void ata_scsi_hotplug(struct work_struct *work);
 extern void ata_schedule_scsi_eh(struct Scsi_Host *shost);
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@ -1238,8 +1238,12 @@ static unsigned int pdc20621_prog_dimm_global(struct ata_host *host)
 	readl(mmio + PDC_SDRAM_CONTROL);

 	/* Turn on for ECC */
-	pdc20621_i2c_read(host, PDC_DIMM0_SPD_DEV_ADDRESS,
-			  PDC_DIMM_SPD_TYPE, &spd0);
+	if (!pdc20621_i2c_read(host, PDC_DIMM0_SPD_DEV_ADDRESS,
+			       PDC_DIMM_SPD_TYPE, &spd0)) {
+		pr_err("Failed in i2c read: device=%#x, subaddr=%#x\n",
+		       PDC_DIMM0_SPD_DEV_ADDRESS, PDC_DIMM_SPD_TYPE);
+		return 1;
+	}
 	if (spd0 == 0x02) {
 		data |= (0x01 << 16);
 		writel(data, mmio + PDC_SDRAM_CONTROL);
@ -1380,8 +1384,12 @@ static unsigned int pdc20621_dimm_init(struct ata_host *host)

 	/* ECC initiliazation. */

-	pdc20621_i2c_read(host, PDC_DIMM0_SPD_DEV_ADDRESS,
-			  PDC_DIMM_SPD_TYPE, &spd0);
+	if (!pdc20621_i2c_read(host, PDC_DIMM0_SPD_DEV_ADDRESS,
+			       PDC_DIMM_SPD_TYPE, &spd0)) {
+		pr_err("Failed in i2c read: device=%#x, subaddr=%#x\n",
+		       PDC_DIMM0_SPD_DEV_ADDRESS, PDC_DIMM_SPD_TYPE);
+		return 1;
+	}
 	if (spd0 == 0x02) {
 		void *buf;
 		VPRINTK("Start ECC initialization\n");
--- a/drivers/base/regmap/regcache-rbtree.c
+++ b/drivers/base/regmap/regcache-rbtree.c
@ -296,11 +296,20 @@ static int regcache_rbtree_insert_to_block(struct regmap *map,
 	if (!blk)
 		return -ENOMEM;

-	present = krealloc(rbnode->cache_present,
-		    BITS_TO_LONGS(blklen) * sizeof(*present), GFP_KERNEL);
-	if (!present) {
-		kfree(blk);
-		return -ENOMEM;
+	if (BITS_TO_LONGS(blklen) > BITS_TO_LONGS(rbnode->blklen)) {
+		present = krealloc(rbnode->cache_present,
+				   BITS_TO_LONGS(blklen) * sizeof(*present),
+				   GFP_KERNEL);
+		if (!present) {
+			kfree(blk);
+			return -ENOMEM;
+		}
+
+		memset(present + BITS_TO_LONGS(rbnode->blklen), 0,
+		       (BITS_TO_LONGS(blklen) - BITS_TO_LONGS(rbnode->blklen))
+		       * sizeof(*present));
+	} else {
+		present = rbnode->cache_present;
 	}

 	/* insert the register value in the correct place in the rbnode block */
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@ -369,8 +369,8 @@ static void purge_persistent_gnt(struct xen_blkif *blkif)
 		return;
 	}

-	if (work_pending(&blkif->persistent_purge_work)) {
-		pr_alert_ratelimited("Scheduled work from previous purge is still pending, cannot purge list\n");
+	if (work_busy(&blkif->persistent_purge_work)) {
+		pr_alert_ratelimited("Scheduled work from previous purge is still busy, cannot purge list\n");
 		return;
 	}

--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@ -179,6 +179,7 @@ static DEFINE_SPINLOCK(minor_lock);
 	((_segs + SEGS_PER_INDIRECT_FRAME - 1)/SEGS_PER_INDIRECT_FRAME)

 static int blkfront_setup_indirect(struct blkfront_info *info);
+static int blkfront_gather_backend_features(struct blkfront_info *info);

 static int get_id_from_freelist(struct blkfront_info *info)
 {
@ -1128,8 +1129,10 @@ static void blkif_completion(struct blk_shadow *s, struct blkfront_info *info,
 				 * Add the used indirect page back to the list of
 				 * available pages for indirect grefs.
 				 */
-				indirect_page = pfn_to_page(s->indirect_grants[i]->pfn);
-				list_add(&indirect_page->lru, &info->indirect_pages);
+				if (!info->feature_persistent) {
+					indirect_page = pfn_to_page(s->indirect_grants[i]->pfn);
+					list_add(&indirect_page->lru, &info->indirect_pages);
+				}
 				s->indirect_grants[i]->gref = GRANT_INVALID_REF;
 				list_add_tail(&s->indirect_grants[i]->node, &info->grants);
 			}
@ -1519,7 +1522,7 @@ static int blkif_recover(struct blkfront_info *info)
 	info->shadow_free = info->ring.req_prod_pvt;
 	info->shadow[BLK_RING_SIZE(info)-1].req.u.rw.id = 0x0fffffff;

-	rc = blkfront_setup_indirect(info);
+	rc = blkfront_gather_backend_features(info);
 	if (rc) {
 		kfree(copy);
 		return rc;
@ -1720,20 +1723,13 @@ static void blkfront_setup_discard(struct blkfront_info *info)

 static int blkfront_setup_indirect(struct blkfront_info *info)
 {
-	unsigned int indirect_segments, segs;
+	unsigned int segs;
 	int err, i;

-	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
-			    "feature-max-indirect-segments", "%u", &indirect_segments,
-			    NULL);
-	if (err) {
-		info->max_indirect_segments = 0;
+	if (info->max_indirect_segments == 0)
 		segs = BLKIF_MAX_SEGMENTS_PER_REQUEST;
-	} else {
-		info->max_indirect_segments = min(indirect_segments,
-						  xen_blkif_max_segments);
+	else
 		segs = info->max_indirect_segments;
-	}

 	err = fill_grant_buffer(info, (segs + INDIRECT_GREFS(segs)) * BLK_RING_SIZE(info));
 	if (err)
@ -1796,6 +1792,68 @@ static int blkfront_setup_indirect(struct blkfront_info *info)
 	return -ENOMEM;
 }

+/*
+ * Gather all backend feature-*
+ */
+static int blkfront_gather_backend_features(struct blkfront_info *info)
+{
+	int err;
+	int barrier, flush, discard, persistent;
+	unsigned int indirect_segments;
+
+	info->feature_flush = 0;
+
+	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
+			"feature-barrier", "%d", &barrier,
+			NULL);
+
+	/*
+	 * If there's no "feature-barrier" defined, then it means
+	 * we're dealing with a very old backend which writes
+	 * synchronously; nothing to do.
+	 *
+	 * If there are barriers, then we use flush.
+	 */
+	if (!err && barrier)
+		info->feature_flush = REQ_FLUSH | REQ_FUA;
+	/*
+	 * And if there is "feature-flush-cache" use that above
+	 * barriers.
+	 */
+	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
+			"feature-flush-cache", "%d", &flush,
+			NULL);
+
+	if (!err && flush)
+		info->feature_flush = REQ_FLUSH;
+
+	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
+			"feature-discard", "%d", &discard,
+			NULL);
+
+	if (!err && discard)
+		blkfront_setup_discard(info);
+
+	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
+			"feature-persistent", "%u", &persistent,
+			NULL);
+	if (err)
+		info->feature_persistent = 0;
+	else
+		info->feature_persistent = persistent;
+
+	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
+			    "feature-max-indirect-segments", "%u", &indirect_segments,
+			    NULL);
+	if (err)
+		info->max_indirect_segments = 0;
+	else
+		info->max_indirect_segments = min(indirect_segments,
+						  xen_blkif_max_segments);
+
+	return blkfront_setup_indirect(info);
+}
+
 /*
 * Invoked when the backend is finally 'ready' (and has told produced
 * the details about the physical device - #sectors, size, etc).
@ -1807,7 +1865,6 @@ static void blkfront_connect(struct blkfront_info *info)
 	unsigned int physical_sector_size;
 	unsigned int binfo;
 	int err;
-	int barrier, flush, discard, persistent;

 	switch (info->connected) {
 	case BLKIF_STATE_CONNECTED:
@ -1864,48 +1921,7 @@ static void blkfront_connect(struct blkfront_info *info)
 	if (err != 1)
 		physical_sector_size = sector_size;

-	info->feature_flush = 0;
-
-	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
-			    "feature-barrier", "%d", &barrier,
-			    NULL);
-
-	/*
-	 * If there's no "feature-barrier" defined, then it means
-	 * we're dealing with a very old backend which writes
-	 * synchronously; nothing to do.
-	 *
-	 * If there are barriers, then we use flush.
-	 */
-	if (!err && barrier)
-		info->feature_flush = REQ_FLUSH | REQ_FUA;
-	/*
-	 * And if there is "feature-flush-cache" use that above
-	 * barriers.
-	 */
-	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
-			    "feature-flush-cache", "%d", &flush,
-			    NULL);
-
-	if (!err && flush)
-		info->feature_flush = REQ_FLUSH;
-
-	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
-			    "feature-discard", "%d", &discard,
-			    NULL);
-
-	if (!err && discard)
-		blkfront_setup_discard(info);
-
-	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
-			    "feature-persistent", "%u", &persistent,
-			    NULL);
-	if (err)
-		info->feature_persistent = 0;
-	else
-		info->feature_persistent = persistent;
-
-	err = blkfront_setup_indirect(info);
+	err = blkfront_gather_backend_features(info);
 	if (err) {
 		xenbus_dev_fatal(info->xbdev, err, "setup_indirect at %s",
 				 info->xbdev->otherend);
--- a/drivers/block/zram/zram_drv.c
+++ b/drivers/block/zram/zram_drv.c
@ -496,10 +496,9 @@ static void zram_meta_free(struct zram_meta *meta, u64 disksize)
 	kfree(meta);
 }

-static struct zram_meta *zram_meta_alloc(int device_id, u64 disksize)
+static struct zram_meta *zram_meta_alloc(char *pool_name, u64 disksize)
 {
 	size_t num_pages;
-	char pool_name[8];
 	struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL);

 	if (!meta)
@ -512,7 +511,6 @@ static struct zram_meta *zram_meta_alloc(int device_id, u64 disksize)
 		goto out_error;
 	}

-	snprintf(pool_name, sizeof(pool_name), "zram%d", device_id);
 	meta->mem_pool = zs_create_pool(pool_name, GFP_NOIO | __GFP_HIGHMEM);
 	if (!meta->mem_pool) {
 		pr_err("Error creating memory pool\n");
@ -1031,7 +1029,7 @@ static ssize_t disksize_store(struct device *dev,
 		return -EINVAL;

 	disksize = PAGE_ALIGN(disksize);
-	meta = zram_meta_alloc(zram->disk->first_minor, disksize);
+	meta = zram_meta_alloc(zram->disk->disk_name, disksize);
 	if (!meta)
 		return -ENOMEM;

--- a/drivers/clk/pxa/clk-pxa3xx.c
+++ b/drivers/clk/pxa/clk-pxa3xx.c
@ -126,7 +126,7 @@ PARENTS(pxa3xx_ac97_bus) = { "ring_osc_60mhz", "ac97" };
 PARENTS(pxa3xx_sbus) = { "ring_osc_60mhz", "system_bus" };
 PARENTS(pxa3xx_smemcbus) = { "ring_osc_60mhz", "smemc" };

-#define CKEN_AB(bit) ((CKEN_ ## bit > 31) ? &CKENA : &CKENB)
+#define CKEN_AB(bit) ((CKEN_ ## bit > 31) ? &CKENB : &CKENA)
 #define PXA3XX_CKEN(dev_id, con_id, parents, mult_lp, div_lp, mult_hp,	\
 		    div_hp, bit, is_lp, flags)				\
 	PXA_CKEN(dev_id, con_id, bit, parents, mult_lp, div_lp,		\
--- a/drivers/clocksource/sh_cmt.c
+++ b/drivers/clocksource/sh_cmt.c
@ -661,6 +661,9 @@ static void sh_cmt_clocksource_suspend(struct clocksource *cs)
 {
 	struct sh_cmt_channel *ch = cs_to_sh_cmt(cs);

+	if (!ch->cs_enabled)
+		return;
+
 	sh_cmt_stop(ch, FLAG_CLOCKSOURCE);
 	pm_genpd_syscore_poweroff(&ch->cmt->pdev->dev);
 }
@ -669,6 +672,9 @@ static void sh_cmt_clocksource_resume(struct clocksource *cs)
 {
 	struct sh_cmt_channel *ch = cs_to_sh_cmt(cs);

+	if (!ch->cs_enabled)
+		return;
+
 	pm_genpd_syscore_poweron(&ch->cmt->pdev->dev);
 	sh_cmt_start(ch, FLAG_CLOCKSOURCE);
 }
--- a/drivers/clocksource/timer-imx-gpt.c
+++ b/drivers/clocksource/timer-imx-gpt.c
@ -462,6 +462,7 @@ void __init mxc_timer_init(unsigned long pbase, int irq, enum imx_gpt_type type)
 	BUG_ON(!imxtm->base);

 	imxtm->type = type;
+	imxtm->irq = irq;

 	_mxc_timer_init(imxtm);
 }
--- a/drivers/cpufreq/exynos-cpufreq.c
+++ b/drivers/cpufreq/exynos-cpufreq.c
@ -180,7 +180,7 @@ static int exynos_cpufreq_probe(struct platform_device *pdev)
 		ret = exynos5250_cpufreq_init(exynos_info);
 	} else {
 		pr_err("%s: Unknown SoC type\n", __func__);
-		return -ENODEV;
+		ret = -ENODEV;
 	}

 	if (ret)
@ -188,12 +188,14 @@ static int exynos_cpufreq_probe(struct platform_device *pdev)

 	if (exynos_info->set_freq == NULL) {
 		dev_err(&pdev->dev, "No set_freq function (ERR)\n");
+		ret = -EINVAL;
 		goto err_vdd_arm;
 	}

 	arm_regulator = regulator_get(NULL, "vdd_arm");
 	if (IS_ERR(arm_regulator)) {
 		dev_err(&pdev->dev, "failed to get resource vdd_arm\n");
+		ret = -EINVAL;
 		goto err_vdd_arm;
 	}

@ -225,7 +227,7 @@ static int exynos_cpufreq_probe(struct platform_device *pdev)
 	regulator_put(arm_regulator);
 err_vdd_arm:
 	kfree(exynos_info);
-	return -EINVAL;
+	return ret;
 }

 static struct platform_driver exynos_cpufreq_platdrv = {
--- a/drivers/crypto/caam/caamhash.c
+++ b/drivers/crypto/caam/caamhash.c
@ -909,13 +909,14 @@ static int ahash_final_ctx(struct ahash_request *req)
 			  state->buflen_1;
 	u32 *sh_desc = ctx->sh_desc_fin, *desc;
 	dma_addr_t ptr = ctx->sh_desc_fin_dma;
-	int sec4_sg_bytes;
+	int sec4_sg_bytes, sec4_sg_src_index;
 	int digestsize = crypto_ahash_digestsize(ahash);
 	struct ahash_edesc *edesc;
 	int ret = 0;
 	int sh_len;

-	sec4_sg_bytes = (1 + (buflen ? 1 : 0)) * sizeof(struct sec4_sg_entry);
+	sec4_sg_src_index = 1 + (buflen ? 1 : 0);
+	sec4_sg_bytes = sec4_sg_src_index * sizeof(struct sec4_sg_entry);

 	/* allocate space for base edesc and hw desc commands, link tables */
 	edesc = kmalloc(sizeof(struct ahash_edesc) + DESC_JOB_IO_LEN +
@ -942,7 +943,7 @@ static int ahash_final_ctx(struct ahash_request *req)
 	state->buf_dma = try_buf_map_to_sec4_sg(jrdev, edesc->sec4_sg + 1,
 						buf, state->buf_dma, buflen,
 						last_buflen);
-	(edesc->sec4_sg + sec4_sg_bytes - 1)->len |= SEC4_SG_LEN_FIN;
+	(edesc->sec4_sg + sec4_sg_src_index - 1)->len |= SEC4_SG_LEN_FIN;

 	edesc->sec4_sg_dma = dma_map_single(jrdev, edesc->sec4_sg,
 					    sec4_sg_bytes, DMA_TO_DEVICE);
--- a/drivers/crypto/nx/nx-sha256.c
+++ b/drivers/crypto/nx/nx-sha256.c
@ -71,7 +71,6 @@ static int nx_sha256_update(struct shash_desc *desc, const u8 *data,
 	struct sha256_state *sctx = shash_desc_ctx(desc);
 	struct nx_crypto_ctx *nx_ctx = crypto_tfm_ctx(&desc->tfm->base);
 	struct nx_csbcpb *csbcpb = (struct nx_csbcpb *)nx_ctx->csbcpb;
-	struct nx_sg *in_sg;
 	struct nx_sg *out_sg;
 	u64 to_process = 0, leftover, total;
 	unsigned long irq_flags;
@ -97,7 +96,6 @@ static int nx_sha256_update(struct shash_desc *desc, const u8 *data,
 	NX_CPB_FDM(csbcpb) |= NX_FDM_INTERMEDIATE;
 	NX_CPB_FDM(csbcpb) |= NX_FDM_CONTINUATION;

-	in_sg = nx_ctx->in_sg;
 	max_sg_len = min_t(u64, nx_ctx->ap->sglen,
 			nx_driver.of.max_sg_len/sizeof(struct nx_sg));
 	max_sg_len = min_t(u64, max_sg_len,
@ -114,17 +112,12 @@ static int nx_sha256_update(struct shash_desc *desc, const u8 *data,
 	}

 	do {
-		/*
-		 * to_process: the SHA256_BLOCK_SIZE data chunk to process in
-		 * this update. This value is also restricted by the sg list
-		 * limits.
-		 */
-		to_process = total - to_process;
-		to_process = to_process & ~(SHA256_BLOCK_SIZE - 1);
+		int used_sgs = 0;
+		struct nx_sg *in_sg = nx_ctx->in_sg;

 		if (buf_len) {
 			data_len = buf_len;
-			in_sg = nx_build_sg_list(nx_ctx->in_sg,
+			in_sg = nx_build_sg_list(in_sg,
 						 (u8 *) sctx->buf,
 						 &data_len,
 						 max_sg_len);
@ -133,15 +126,27 @@ static int nx_sha256_update(struct shash_desc *desc, const u8 *data,
 				rc = -EINVAL;
 				goto out;
 			}
+			used_sgs = in_sg - nx_ctx->in_sg;
 		}

+		/* to_process: SHA256_BLOCK_SIZE aligned chunk to be
+		 * processed in this iteration. This value is restricted
+		 * by sg list limits and number of sgs we already used
+		 * for leftover data. (see above)
+		 * In ideal case, we could allow NX_PAGE_SIZE * max_sg_len,
+		 * but because data may not be aligned, we need to account
+		 * for that too. */
+		to_process = min_t(u64, total,
+			(max_sg_len - 1 - used_sgs) * NX_PAGE_SIZE);
+		to_process = to_process & ~(SHA256_BLOCK_SIZE - 1);
+
 		data_len = to_process - buf_len;
 		in_sg = nx_build_sg_list(in_sg, (u8 *) data,
 					 &data_len, max_sg_len);

 		nx_ctx->op.inlen = (nx_ctx->in_sg - in_sg) * sizeof(struct nx_sg);

-		to_process = (data_len + buf_len);
+		to_process = data_len + buf_len;
 		leftover = total - to_process;

 		/*
--- a/drivers/crypto/nx/nx-sha512.c
+++ b/drivers/crypto/nx/nx-sha512.c
@ -71,7 +71,6 @@ static int nx_sha512_update(struct shash_desc *desc, const u8 *data,
 	struct sha512_state *sctx = shash_desc_ctx(desc);
 	struct nx_crypto_ctx *nx_ctx = crypto_tfm_ctx(&desc->tfm->base);
 	struct nx_csbcpb *csbcpb = (struct nx_csbcpb *)nx_ctx->csbcpb;
-	struct nx_sg *in_sg;
 	struct nx_sg *out_sg;
 	u64 to_process, leftover = 0, total;
 	unsigned long irq_flags;
@ -97,7 +96,6 @@ static int nx_sha512_update(struct shash_desc *desc, const u8 *data,
 	NX_CPB_FDM(csbcpb) |= NX_FDM_INTERMEDIATE;
 	NX_CPB_FDM(csbcpb) |= NX_FDM_CONTINUATION;

-	in_sg = nx_ctx->in_sg;
 	max_sg_len = min_t(u64, nx_ctx->ap->sglen,
 			nx_driver.of.max_sg_len/sizeof(struct nx_sg));
 	max_sg_len = min_t(u64, max_sg_len,
@ -114,18 +112,12 @@ static int nx_sha512_update(struct shash_desc *desc, const u8 *data,
 	}

 	do {
-		/*
-		 * to_process: the SHA512_BLOCK_SIZE data chunk to process in
-		 * this update. This value is also restricted by the sg list
-		 * limits.
-		 */
-		to_process = total - leftover;
-		to_process = to_process & ~(SHA512_BLOCK_SIZE - 1);
-		leftover = total - to_process;
+		int used_sgs = 0;
+		struct nx_sg *in_sg = nx_ctx->in_sg;

 		if (buf_len) {
 			data_len = buf_len;
-			in_sg = nx_build_sg_list(nx_ctx->in_sg,
+			in_sg = nx_build_sg_list(in_sg,
 						 (u8 *) sctx->buf,
 						 &data_len, max_sg_len);

@ -133,8 +125,20 @@ static int nx_sha512_update(struct shash_desc *desc, const u8 *data,
 				rc = -EINVAL;
 				goto out;
 			}
+			used_sgs = in_sg - nx_ctx->in_sg;
 		}

+		/* to_process: SHA512_BLOCK_SIZE aligned chunk to be
+		 * processed in this iteration. This value is restricted
+		 * by sg list limits and number of sgs we already used
+		 * for leftover data. (see above)
+		 * In ideal case, we could allow NX_PAGE_SIZE * max_sg_len,
+		 * but because data may not be aligned, we need to account
+		 * for that too. */
+		to_process = min_t(u64, total,
+			(max_sg_len - 1 - used_sgs) * NX_PAGE_SIZE);
+		to_process = to_process & ~(SHA512_BLOCK_SIZE - 1);
+
 		data_len = to_process - buf_len;
 		in_sg = nx_build_sg_list(in_sg, (u8 *) data,
 					 &data_len, max_sg_len);
@ -146,7 +150,7 @@ static int nx_sha512_update(struct shash_desc *desc, const u8 *data,
 			goto out;
 		}

-		to_process = (data_len + buf_len);
+		to_process = data_len + buf_len;
 		leftover = total - to_process;

 		/*
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@ -689,6 +689,10 @@ struct dma_chan *dma_request_slave_channel(struct device *dev,
 	struct dma_chan *ch = dma_request_slave_channel_reason(dev, name);
 	if (IS_ERR(ch))
 		return NULL;
+
+	dma_cap_set(DMA_PRIVATE, ch->device->cap_mask);
+	ch->device->privatecnt++;
+
 	return ch;
 }
 EXPORT_SYMBOL_GPL(dma_request_slave_channel);
--- a/drivers/edac/ppc4xx_edac.c
+++ b/drivers/edac/ppc4xx_edac.c
@ -920,7 +920,7 @@ static int ppc4xx_edac_init_csrows(struct mem_ctl_info *mci, u32 mcopt1)
 	 */

 	for (row = 0; row < mci->nr_csrows; row++) {
-		struct csrow_info *csi = &mci->csrows[row];
+		struct csrow_info *csi = mci->csrows[row];

 		/*
 		 * Get the configuration settings for this
--- a/drivers/firmware/broadcom/bcm47xx_nvram.c
+++ b/drivers/firmware/broadcom/bcm47xx_nvram.c
@ -245,4 +245,4 @@ char *bcm47xx_nvram_get_contents(size_t *nvram_size)
 }
 EXPORT_SYMBOL(bcm47xx_nvram_get_contents);

-MODULE_LICENSE("GPLv2");
+MODULE_LICENSE("GPL v2");
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@ -37,9 +37,29 @@ config DRM_KMS_FB_HELPER
 	select FB
 	select FRAMEBUFFER_CONSOLE if !EXPERT
 	select FRAMEBUFFER_CONSOLE_DETECT_PRIMARY if FRAMEBUFFER_CONSOLE
+	select FB_SYS_FOPS
+	select FB_SYS_FILLRECT
+	select FB_SYS_COPYAREA
+	select FB_SYS_IMAGEBLIT
+	select FB_CFB_FILLRECT
+	select FB_CFB_COPYAREA
+	select FB_CFB_IMAGEBLIT
 	help
 	  FBDEV helpers for KMS drivers.

+config DRM_FBDEV_EMULATION
+	bool "Enable legacy fbdev support for your modesetting driver"
+	depends on DRM
+	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
+	default y
+	help
+	  Choose this option if you have a need for the legacy fbdev
+	  support. Note that this support also provides the linux console
+	  support on top of your modesetting driver.
+
+	  If in doubt, say "Y".
+
 config DRM_LOAD_EDID_FIRMWARE
 	bool "Allow to specify an EDID data set instead of probing for it"
 	depends on DRM_KMS_HELPER
@ -79,8 +99,6 @@ config DRM_KMS_CMA_HELPER

 source "drivers/gpu/drm/i2c/Kconfig"

-source "drivers/gpu/drm/bridge/Kconfig"
-
 config DRM_TDFX
 	tristate "3dfx Banshee/Voodoo3+"
 	depends on DRM && PCI
@ -110,6 +128,7 @@ config DRM_RADEON
 	select POWER_SUPPLY
 	select HWMON
 	select BACKLIGHT_CLASS_DEVICE
+	select BACKLIGHT_LCD_SUPPORT
 	select INTERVAL_TREE
 	help
 	  Choose this option if you have an ATI Radeon graphics card.  There
@ -133,6 +152,7 @@ config DRM_AMDGPU
 	select POWER_SUPPLY
 	select HWMON
 	select BACKLIGHT_CLASS_DEVICE
+	select BACKLIGHT_LCD_SUPPORT
 	select INTERVAL_TREE
 	help
 	  Choose this option if you have a recent AMD Radeon graphics card.
@ -231,10 +251,14 @@ source "drivers/gpu/drm/virtio/Kconfig"

 source "drivers/gpu/drm/msm/Kconfig"

+source "drivers/gpu/drm/fsl-dcu/Kconfig"
+
 source "drivers/gpu/drm/tegra/Kconfig"

 source "drivers/gpu/drm/panel/Kconfig"

+source "drivers/gpu/drm/bridge/Kconfig"
+
 source "drivers/gpu/drm/sti/Kconfig"

 source "drivers/gpu/drm/amd/amdkfd/Kconfig"
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@ -23,7 +23,7 @@ drm-$(CONFIG_OF) += drm_of.o
 drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o \
 		drm_plane_helper.o drm_dp_mst_topology.o drm_atomic_helper.o
 drm_kms_helper-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o
-drm_kms_helper-$(CONFIG_DRM_KMS_FB_HELPER) += drm_fb_helper.o
+drm_kms_helper-$(CONFIG_DRM_FBDEV_EMULATION) += drm_fb_helper.o
 drm_kms_helper-$(CONFIG_DRM_KMS_CMA_HELPER) += drm_fb_cma_helper.o

 obj-$(CONFIG_DRM_KMS_HELPER) += drm_kms_helper.o
@ -70,3 +70,4 @@ obj-$(CONFIG_DRM_IMX) += imx/
 obj-y			+= i2c/
 obj-y			+= panel/
 obj-y			+= bridge/
+obj-$(CONFIG_DRM_FSL_DCU) += fsl-dcu/
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@ -3,7 +3,9 @@
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.

 ccflags-y := -Iinclude/drm -Idrivers/gpu/drm/amd/include/asic_reg \
-	-Idrivers/gpu/drm/amd/include
+	-Idrivers/gpu/drm/amd/include \
+	-Idrivers/gpu/drm/amd/amdgpu \
+	-Idrivers/gpu/drm/amd/scheduler

 amdgpu-y := amdgpu_drv.o

@ -21,7 +23,8 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \

 # add asic specific block
 amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o gmc_v7_0.o cik_ih.o kv_smc.o kv_dpm.o \
-	ci_smc.o ci_dpm.o dce_v8_0.o gfx_v7_0.o cik_sdma.o uvd_v4_2.o vce_v2_0.o
+	ci_smc.o ci_dpm.o dce_v8_0.o gfx_v7_0.o cik_sdma.o uvd_v4_2.o vce_v2_0.o \
+	amdgpu_amdkfd_gfx_v7.o

 amdgpu-y += \
 	vi.o
@ -43,6 +46,7 @@ amdgpu-y += \
 	amdgpu_dpm.o \
 	cz_smc.o cz_dpm.o \
 	tonga_smc.o tonga_dpm.o \
+	fiji_smc.o fiji_dpm.o \
 	iceland_smc.o iceland_dpm.o

 # add DCE block
@ -74,9 +78,17 @@ amdgpu-y += \
 # add amdkfd interfaces
 amdgpu-y += \
 	 amdgpu_amdkfd.o \
-	 amdgpu_amdkfd_gfx_v7.o \
 	 amdgpu_amdkfd_gfx_v8.o

+# add cgs
+amdgpu-y += amdgpu_cgs.o
+
+# GPU scheduler
+amdgpu-y += \
+	../scheduler/gpu_scheduler.o \
+	../scheduler/sched_fence.o \
+	amdgpu_sched.o
+
 amdgpu-$(CONFIG_COMPAT) += amdgpu_ioc32.o
 amdgpu-$(CONFIG_VGA_SWITCHEROO) += amdgpu_atpx_handler.o
 amdgpu-$(CONFIG_ACPI) += amdgpu_acpi.o
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@ -42,17 +42,19 @@
 #include <ttm/ttm_module.h>
 #include <ttm/ttm_execbuf_util.h>

+#include <drm/drmP.h>
 #include <drm/drm_gem.h>
 #include <drm/amdgpu_drm.h>

 #include "amd_shared.h"
-#include "amdgpu_family.h"
 #include "amdgpu_mode.h"
 #include "amdgpu_ih.h"
 #include "amdgpu_irq.h"
 #include "amdgpu_ucode.h"
 #include "amdgpu_gds.h"

+#include "gpu_scheduler.h"
+
 /*
 * Modules parameters.
 */
@ -77,7 +79,11 @@ extern int amdgpu_bapm;
 extern int amdgpu_deep_color;
 extern int amdgpu_vm_size;
 extern int amdgpu_vm_block_size;
+extern int amdgpu_enable_scheduler;
+extern int amdgpu_sched_jobs;
+extern int amdgpu_sched_hw_submission;

+#define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS	        3000
 #define AMDGPU_MAX_USEC_TIMEOUT			100000	/* 100 ms */
 #define AMDGPU_FENCE_JIFFIES_TIMEOUT		(HZ / 2)
 /* AMDGPU_IB_POOL_SIZE must be a power of 2 */
@ -178,6 +184,7 @@ struct amdgpu_ring;
 struct amdgpu_semaphore;
 struct amdgpu_cs_parser;
 struct amdgpu_irq_src;
+struct amdgpu_fpriv;

 enum amdgpu_cp_irq {
 	AMDGPU_CP_IRQ_GFX_EOP = 0,
@ -381,10 +388,10 @@ struct amdgpu_fence_driver {
 	uint64_t			sync_seq[AMDGPU_MAX_RINGS];
 	atomic64_t			last_seq;
 	bool				initialized;
-	bool				delayed_irq;
 	struct amdgpu_irq_src		*irq_src;
 	unsigned			irq_type;
 	struct delayed_work             lockup_work;
+	wait_queue_head_t		fence_queue;
 };

 /* some special values for the owner field */
@ -423,20 +430,18 @@ void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring);
 int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
 				   struct amdgpu_irq_src *irq_src,
 				   unsigned irq_type);
+void amdgpu_fence_driver_suspend(struct amdgpu_device *adev);
+void amdgpu_fence_driver_resume(struct amdgpu_device *adev);
 int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
 		      struct amdgpu_fence **fence);
-int amdgpu_fence_recreate(struct amdgpu_ring *ring, void *owner,
-			  uint64_t seq, struct amdgpu_fence **fence);
 void amdgpu_fence_process(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_next(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);
 unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring);

-bool amdgpu_fence_signaled(struct amdgpu_fence *fence);
-int amdgpu_fence_wait(struct amdgpu_fence *fence, bool interruptible);
-int amdgpu_fence_wait_any(struct amdgpu_device *adev,
+signed long amdgpu_fence_wait_any(struct amdgpu_device *adev,
 			  struct amdgpu_fence **fences,
-			  bool intr);
+			  bool intr, long t);
 struct amdgpu_fence *amdgpu_fence_ref(struct amdgpu_fence *fence);
 void amdgpu_fence_unref(struct amdgpu_fence **fence);

@ -481,7 +486,7 @@ static inline bool amdgpu_fence_is_earlier(struct amdgpu_fence *a,
 	return a->seq < b->seq;
 }

-int amdgpu_user_fence_emit(struct amdgpu_ring *ring, struct amdgpu_user_fence *user, 
+int amdgpu_user_fence_emit(struct amdgpu_ring *ring, struct amdgpu_user_fence *user,
 			   void *owner, struct amdgpu_fence **fence);

 /*
@ -532,14 +537,16 @@ struct amdgpu_bo_va_mapping {
 struct amdgpu_bo_va {
 	/* protected by bo being reserved */
 	struct list_head		bo_list;
-	uint64_t			addr;
-	struct amdgpu_fence		*last_pt_update;
+	struct fence		        *last_pt_update;
 	unsigned			ref_count;

-	/* protected by vm mutex */
-	struct list_head		mappings;
+	/* protected by vm mutex and spinlock */
 	struct list_head		vm_status;

+	/* mappings for this bo_va */
+	struct list_head		invalids;
+	struct list_head		valids;
+
 	/* constant after initialization */
 	struct amdgpu_vm		*vm;
 	struct amdgpu_bo		*bo;
@ -697,8 +704,8 @@ struct amdgpu_sync {
 };

 void amdgpu_sync_create(struct amdgpu_sync *sync);
-void amdgpu_sync_fence(struct amdgpu_sync *sync,
-		       struct amdgpu_fence *fence);
+int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
+		      struct fence *f);
 int amdgpu_sync_resv(struct amdgpu_device *adev,
 		     struct amdgpu_sync *sync,
 		     struct reservation_object *resv,
@ -821,7 +828,9 @@ struct amdgpu_flip_work {
 	uint64_t			base;
 	struct drm_pending_vblank_event *event;
 	struct amdgpu_bo		*old_rbo;
-	struct fence			*fence;
+	struct fence			*excl;
+	unsigned			shared_count;
+	struct fence			**shared;
 };


@ -844,6 +853,8 @@ struct amdgpu_ib {
 	uint32_t			gws_base, gws_size;
 	uint32_t			oa_base, oa_size;
 	uint32_t			flags;
+	/* resulting sequence number */
+	uint64_t			sequence;
 };

 enum amdgpu_ring_type {
@ -854,11 +865,23 @@ enum amdgpu_ring_type {
 	AMDGPU_RING_TYPE_VCE
 };

+extern struct amd_sched_backend_ops amdgpu_sched_ops;
+
+int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev,
+					 struct amdgpu_ring *ring,
+					 struct amdgpu_ib *ibs,
+					 unsigned num_ibs,
+					 int (*free_job)(struct amdgpu_cs_parser *),
+					 void *owner,
+					 struct fence **fence);
+
 struct amdgpu_ring {
 	struct amdgpu_device		*adev;
 	const struct amdgpu_ring_funcs	*funcs;
 	struct amdgpu_fence_driver	fence_drv;
+	struct amd_gpu_scheduler 	*scheduler;

+	spinlock_t              fence_lock;
 	struct mutex		*ring_lock;
 	struct amdgpu_bo	*ring_obj;
 	volatile uint32_t	*ring;
@ -892,6 +915,7 @@ struct amdgpu_ring {
 	struct amdgpu_ctx	*current_ctx;
 	enum amdgpu_ring_type	type;
 	char			name[16];
+	bool                    is_pte_ring;
 };

 /*
@ -943,18 +967,22 @@ struct amdgpu_vm {

 	struct rb_root		va;

-	/* protecting invalidated and freed */
+	/* protecting invalidated */
 	spinlock_t		status_lock;

 	/* BOs moved, but not yet updated in the PT */
 	struct list_head	invalidated;

-	/* BOs freed, but not yet updated in the PT */
+	/* BOs cleared in the PT because of a move */
+	struct list_head	cleared;
+
+	/* BO mappings freed, but not yet updated in the PT */
 	struct list_head	freed;

 	/* contains the page directory */
 	struct amdgpu_bo	*page_directory;
 	unsigned		max_pde_used;
+	struct fence		*page_directory_fence;

 	/* array of page tables, one for each page directory entry */
 	struct amdgpu_vm_pt	*page_tables;
@ -983,27 +1011,47 @@ struct amdgpu_vm_manager {
 * context related structures
 */

-struct amdgpu_ctx_state {
-	uint64_t flags;
-	uint32_t hangs;
+#define AMDGPU_CTX_MAX_CS_PENDING	16
+
+struct amdgpu_ctx_ring {
+	uint64_t		sequence;
+	struct fence		*fences[AMDGPU_CTX_MAX_CS_PENDING];
+	struct amd_sched_entity	entity;
 };

 struct amdgpu_ctx {
-	/* call kref_get()before CS start and kref_put() after CS fence signaled */
-	struct kref refcount;
-	struct amdgpu_fpriv *fpriv;
-	struct amdgpu_ctx_state state;
-	uint32_t id;
-	unsigned reset_counter;
+	struct kref		refcount;
+	struct amdgpu_device    *adev;
+	unsigned		reset_counter;
+	spinlock_t		ring_lock;
+	struct amdgpu_ctx_ring	rings[AMDGPU_MAX_RINGS];
 };

 struct amdgpu_ctx_mgr {
-	struct amdgpu_device *adev;
-	struct idr ctx_handles;
-	/* lock for IDR system */
-	struct mutex lock;
+	struct amdgpu_device	*adev;
+	struct mutex		lock;
+	/* protected by lock */
+	struct idr		ctx_handles;
 };

+int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel,
+		    struct amdgpu_ctx *ctx);
+void amdgpu_ctx_fini(struct amdgpu_ctx *ctx);
+
+struct amdgpu_ctx *amdgpu_ctx_get(struct amdgpu_fpriv *fpriv, uint32_t id);
+int amdgpu_ctx_put(struct amdgpu_ctx *ctx);
+
+uint64_t amdgpu_ctx_add_fence(struct amdgpu_ctx *ctx, struct amdgpu_ring *ring,
+			      struct fence *fence, uint64_t queued_seq);
+struct fence *amdgpu_ctx_get_fence(struct amdgpu_ctx *ctx,
+				   struct amdgpu_ring *ring, uint64_t seq);
+
+int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
+		     struct drm_file *filp);
+
+void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr);
+void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr);
+
 /*
 * file private structure
 */
@ -1012,7 +1060,7 @@ struct amdgpu_fpriv {
 	struct amdgpu_vm	vm;
 	struct mutex		bo_list_lock;
 	struct idr		bo_list_handles;
-	struct amdgpu_ctx_mgr ctx_mgr;
+	struct amdgpu_ctx_mgr	ctx_mgr;
 };

 /*
@ -1029,6 +1077,8 @@ struct amdgpu_bo_list {
 	struct amdgpu_bo_list_entry *array;
 };

+struct amdgpu_bo_list *
+amdgpu_bo_list_clone(struct amdgpu_bo_list *list);
 struct amdgpu_bo_list *
 amdgpu_bo_list_get(struct amdgpu_fpriv *fpriv, int id);
 void amdgpu_bo_list_put(struct amdgpu_bo_list *list);
@ -1205,6 +1255,14 @@ struct amdgpu_cs_parser {

 	/* user fence */
 	struct amdgpu_user_fence uf;
+
+	struct amdgpu_ring *ring;
+	struct mutex job_lock;
+	struct work_struct job_work;
+	int (*prepare_job)(struct amdgpu_cs_parser *sched_job);
+	int (*run_job)(struct amdgpu_cs_parser *sched_job);
+	int (*free_job)(struct amdgpu_cs_parser *sched_job);
+	struct amd_sched_fence *s_fence;
 };

 static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p, uint32_t ib_idx, int idx)
@ -1849,17 +1907,12 @@ struct amdgpu_atcs {
 	struct amdgpu_atcs_functions functions;
 };

-int amdgpu_ctx_alloc(struct amdgpu_device *adev,struct amdgpu_fpriv *fpriv,
-							uint32_t *id,uint32_t flags);
-int amdgpu_ctx_free(struct amdgpu_device *adev, struct amdgpu_fpriv *fpriv,
-						  uint32_t id);
+/*
+ * CGS
+ */
+void *amdgpu_cgs_create_device(struct amdgpu_device *adev);
+void amdgpu_cgs_destroy_device(void *cgs_device);

-void amdgpu_ctx_fini(struct amdgpu_fpriv *fpriv);
-struct amdgpu_ctx *amdgpu_ctx_get(struct amdgpu_fpriv *fpriv, uint32_t id);
-int amdgpu_ctx_put(struct amdgpu_ctx *ctx);
-
-extern int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
-						 struct drm_file *filp);

 /*
 * Core structure, functions and helpers.
@ -1883,7 +1936,7 @@ struct amdgpu_device {
 	struct rw_semaphore		exclusive_lock;

 	/* ASIC */
-	enum amdgpu_asic_type           asic_type;
+	enum amd_asic_type		asic_type;
 	uint32_t			family;
 	uint32_t			rev_id;
 	uint32_t			external_rev_id;
@ -1976,7 +2029,6 @@ struct amdgpu_device {
 	struct amdgpu_irq_src		hpd_irq;

 	/* rings */
-	wait_queue_head_t		fence_queue;
 	unsigned			fence_context;
 	struct mutex			ring_lock;
 	unsigned			num_rings;
@ -2028,6 +2080,9 @@ struct amdgpu_device {

 	/* amdkfd interface */
 	struct kfd_dev          *kfd;
+
+	/* kernel conext for IB submission */
+	struct amdgpu_ctx	kernel_ctx;
 };

 bool amdgpu_device_is_px(struct drm_device *dev);
@ -2215,6 +2270,12 @@ void amdgpu_pci_config_reset(struct amdgpu_device *adev);
 bool amdgpu_card_posted(struct amdgpu_device *adev);
 void amdgpu_update_display_priority(struct amdgpu_device *adev);
 bool amdgpu_boot_test_post_card(struct amdgpu_device *adev);
+struct amdgpu_cs_parser *amdgpu_cs_parser_create(struct amdgpu_device *adev,
+						 struct drm_file *filp,
+						 struct amdgpu_ctx *ctx,
+						 struct amdgpu_ib *ibs,
+						 uint32_t num_ibs);
+
 int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data);
 int amdgpu_cs_get_ring(struct amdgpu_device *adev, u32 ip_type,
 		       u32 ip_instance, u32 ring,
@ -2278,8 +2339,8 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm);
 struct amdgpu_bo_list_entry *amdgpu_vm_get_bos(struct amdgpu_device *adev,
 					  struct amdgpu_vm *vm,
 					  struct list_head *head);
-struct amdgpu_fence *amdgpu_vm_grab_id(struct amdgpu_ring *ring,
-				       struct amdgpu_vm *vm);
+int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
+		      struct amdgpu_sync *sync);
 void amdgpu_vm_flush(struct amdgpu_ring *ring,
 		     struct amdgpu_vm *vm,
 		     struct amdgpu_fence *updates);
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@ -21,7 +21,7 @@
 */

 #include "amdgpu_amdkfd.h"
-#include "amdgpu_family.h"
+#include "amd_shared.h"
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include <linux/module.h>
@ -50,9 +50,11 @@ bool amdgpu_amdkfd_load_interface(struct amdgpu_device *rdev)
 #endif

 	switch (rdev->asic_type) {
+#ifdef CONFIG_DRM_AMDGPU_CIK
 	case CHIP_KAVERI:
 		kfd2kgd = amdgpu_amdkfd_gfx_7_get_functions();
 		break;
+#endif
 	case CHIP_CARRIZO:
 		kfd2kgd = amdgpu_amdkfd_gfx_8_0_get_functions();
 		break;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@ -450,7 +450,7 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,

 	while (true) {
 		temp = RREG32(mmCP_HQD_ACTIVE);
-		if (temp & CP_HQD_ACTIVE__ACTIVE__SHIFT)
+		if (temp & CP_HQD_ACTIVE__ACTIVE_MASK)
 			break;
 		if (timeout == 0) {
 			pr_err("kfd: cp queue preemption time out (%dms)\n",
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
@ -897,7 +897,7 @@ bool amdgpu_atombios_get_asic_ss_info(struct amdgpu_device *adev,
 					if ((id == ASIC_INTERNAL_ENGINE_SS) ||
 					    (id == ASIC_INTERNAL_MEMORY_SS))
 						ss->rate /= 100;
-					if (adev->flags & AMDGPU_IS_APU)
+					if (adev->flags & AMD_IS_APU)
 						amdgpu_atombios_get_igp_ss_overrides(adev, ss, id);
 					return true;
 				}
@ -1058,7 +1058,7 @@ void amdgpu_atombios_set_memory_clock(struct amdgpu_device *adev,
 	SET_MEMORY_CLOCK_PS_ALLOCATION args;
 	int index = GetIndexIntoMasterTable(COMMAND, SetMemoryClock);

-	if (adev->flags & AMDGPU_IS_APU)
+	if (adev->flags & AMD_IS_APU)
 		return;

 	args.ulTargetMemoryClock = cpu_to_le32(mem_clock);	/* 10 khz */
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c
@ -42,7 +42,7 @@ static int amdgpu_benchmark_do_move(struct amdgpu_device *adev, unsigned size,
 		r = amdgpu_copy_buffer(ring, saddr, daddr, size, NULL, &fence);
 		if (r)
 			goto exit_do_move;
-		r = amdgpu_fence_wait(fence, false);
+		r = fence_wait(&fence->base, false);
 		if (r)
 			goto exit_do_move;
 		amdgpu_fence_unref(&fence);
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
@ -48,7 +48,7 @@ static bool igp_read_bios_from_vram(struct amdgpu_device *adev)
 	resource_size_t vram_base;
 	resource_size_t size = 256 * 1024; /* ??? */

-	if (!(adev->flags & AMDGPU_IS_APU))
+	if (!(adev->flags & AMD_IS_APU))
 		if (!amdgpu_card_posted(adev))
 			return false;

@ -184,7 +184,7 @@ static bool amdgpu_atrm_get_bios(struct amdgpu_device *adev)
 	bool found = false;

 	/* ATRM is for the discrete card only */
-	if (adev->flags & AMDGPU_IS_APU)
+	if (adev->flags & AMD_IS_APU)
 		return false;

 	while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_VGA << 8, pdev)) != NULL) {
@ -246,7 +246,7 @@ static inline bool amdgpu_atrm_get_bios(struct amdgpu_device *adev)

 static bool amdgpu_read_disabled_bios(struct amdgpu_device *adev)
 {
-	if (adev->flags & AMDGPU_IS_APU)
+	if (adev->flags & AMD_IS_APU)
 		return igp_read_bios_from_vram(adev);
 	else
 		return amdgpu_asic_read_disabled_bios(adev);
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@ -62,6 +62,39 @@ static int amdgpu_bo_list_create(struct amdgpu_fpriv *fpriv,
 	return 0;
 }

+struct amdgpu_bo_list *
+amdgpu_bo_list_clone(struct amdgpu_bo_list *list)
+{
+	struct amdgpu_bo_list *result;
+	unsigned i;
+
+	result = kmalloc(sizeof(struct amdgpu_bo_list), GFP_KERNEL);
+	if (!result)
+		return NULL;
+
+	result->array = drm_calloc_large(list->num_entries,
+		sizeof(struct amdgpu_bo_list_entry));
+	if (!result->array) {
+		kfree(result);
+		return NULL;
+	}
+
+	mutex_init(&result->lock);
+	result->gds_obj = list->gds_obj;
+	result->gws_obj = list->gws_obj;
+	result->oa_obj = list->oa_obj;
+	result->has_userptr = list->has_userptr;
+	result->num_entries = list->num_entries;
+
+	memcpy(result->array, list->array, list->num_entries *
+	       sizeof(struct amdgpu_bo_list_entry));
+
+	for (i = 0; i < result->num_entries; ++i)
+		amdgpu_bo_ref(result->array[i].robj);
+
+	return result;
+}
+
 static void amdgpu_bo_list_destroy(struct amdgpu_fpriv *fpriv, int id)
 {
 	struct amdgpu_bo_list *list;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
@ -0,0 +1,838 @@
+/*
+ * Copyright 2015 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ *
+ */
+#include <linux/list.h>
+#include <linux/slab.h>
+#include <linux/pci.h>
+#include <drm/drmP.h>
+#include <linux/firmware.h>
+#include <drm/amdgpu_drm.h>
+#include "amdgpu.h"
+#include "cgs_linux.h"
+#include "atom.h"
+#include "amdgpu_ucode.h"
+
+
+struct amdgpu_cgs_device {
+	struct cgs_device base;
+	struct amdgpu_device *adev;
+};
+
+#define CGS_FUNC_ADEV							\
+	struct amdgpu_device *adev =					\
+		((struct amdgpu_cgs_device *)cgs_device)->adev
+
+static int amdgpu_cgs_gpu_mem_info(void *cgs_device, enum cgs_gpu_mem_type type,
+				   uint64_t *mc_start, uint64_t *mc_size,
+				   uint64_t *mem_size)
+{
+	CGS_FUNC_ADEV;
+	switch(type) {
+	case CGS_GPU_MEM_TYPE__VISIBLE_CONTIG_FB:
+	case CGS_GPU_MEM_TYPE__VISIBLE_FB:
+		*mc_start = 0;
+		*mc_size = adev->mc.visible_vram_size;
+		*mem_size = adev->mc.visible_vram_size - adev->vram_pin_size;
+		break;
+	case CGS_GPU_MEM_TYPE__INVISIBLE_CONTIG_FB:
+	case CGS_GPU_MEM_TYPE__INVISIBLE_FB:
+		*mc_start = adev->mc.visible_vram_size;
+		*mc_size = adev->mc.real_vram_size - adev->mc.visible_vram_size;
+		*mem_size = *mc_size;
+		break;
+	case CGS_GPU_MEM_TYPE__GART_CACHEABLE:
+	case CGS_GPU_MEM_TYPE__GART_WRITECOMBINE:
+		*mc_start = adev->mc.gtt_start;
+		*mc_size = adev->mc.gtt_size;
+		*mem_size = adev->mc.gtt_size - adev->gart_pin_size;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int amdgpu_cgs_gmap_kmem(void *cgs_device, void *kmem,
+				uint64_t size,
+				uint64_t min_offset, uint64_t max_offset,
+				cgs_handle_t *kmem_handle, uint64_t *mcaddr)
+{
+	CGS_FUNC_ADEV;
+	int ret;
+	struct amdgpu_bo *bo;
+	struct page *kmem_page = vmalloc_to_page(kmem);
+	int npages = ALIGN(size, PAGE_SIZE) >> PAGE_SHIFT;
+
+	struct sg_table *sg = drm_prime_pages_to_sg(&kmem_page, npages);
+	ret = amdgpu_bo_create(adev, size, PAGE_SIZE, false,
+			       AMDGPU_GEM_DOMAIN_GTT, 0, sg, &bo);
+	if (ret)
+		return ret;
+	ret = amdgpu_bo_reserve(bo, false);
+	if (unlikely(ret != 0))
+		return ret;
+
+	/* pin buffer into GTT */
+	ret = amdgpu_bo_pin_restricted(bo, AMDGPU_GEM_DOMAIN_GTT,
+				       min_offset, max_offset, mcaddr);
+	amdgpu_bo_unreserve(bo);
+
+	*kmem_handle = (cgs_handle_t)bo;
+	return ret;
+}
+
+static int amdgpu_cgs_gunmap_kmem(void *cgs_device, cgs_handle_t kmem_handle)
+{
+	struct amdgpu_bo *obj = (struct amdgpu_bo *)kmem_handle;
+
+	if (obj) {
+		int r = amdgpu_bo_reserve(obj, false);
+		if (likely(r == 0)) {
+			amdgpu_bo_unpin(obj);
+			amdgpu_bo_unreserve(obj);
+		}
+		amdgpu_bo_unref(&obj);
+
+	}
+	return 0;
+}
+
+static int amdgpu_cgs_alloc_gpu_mem(void *cgs_device,
+				    enum cgs_gpu_mem_type type,
+				    uint64_t size, uint64_t align,
+				    uint64_t min_offset, uint64_t max_offset,
+				    cgs_handle_t *handle)
+{
+	CGS_FUNC_ADEV;
+	uint16_t flags = 0;
+	int ret = 0;
+	uint32_t domain = 0;
+	struct amdgpu_bo *obj;
+	struct ttm_placement placement;
+	struct ttm_place place;
+
+	if (min_offset > max_offset) {
+		BUG_ON(1);
+		return -EINVAL;
+	}
+
+	/* fail if the alignment is not a power of 2 */
+	if (((align != 1) && (align & (align - 1)))
+	    || size == 0 || align == 0)
+		return -EINVAL;
+
+
+	switch(type) {
+	case CGS_GPU_MEM_TYPE__VISIBLE_CONTIG_FB:
+	case CGS_GPU_MEM_TYPE__VISIBLE_FB:
+		flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
+		domain = AMDGPU_GEM_DOMAIN_VRAM;
+		if (max_offset > adev->mc.real_vram_size)
+			return -EINVAL;
+		place.fpfn = min_offset >> PAGE_SHIFT;
+		place.lpfn = max_offset >> PAGE_SHIFT;
+		place.flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED |
+			TTM_PL_FLAG_VRAM;
+		break;
+	case CGS_GPU_MEM_TYPE__INVISIBLE_CONTIG_FB:
+	case CGS_GPU_MEM_TYPE__INVISIBLE_FB:
+		flags = AMDGPU_GEM_CREATE_NO_CPU_ACCESS;
+		domain = AMDGPU_GEM_DOMAIN_VRAM;
+		if (adev->mc.visible_vram_size < adev->mc.real_vram_size) {
+			place.fpfn =
+				max(min_offset, adev->mc.visible_vram_size) >> PAGE_SHIFT;
+			place.lpfn =
+				min(max_offset, adev->mc.real_vram_size) >> PAGE_SHIFT;
+			place.flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED |
+				TTM_PL_FLAG_VRAM;
+		}
+
+		break;
+	case CGS_GPU_MEM_TYPE__GART_CACHEABLE:
+		domain = AMDGPU_GEM_DOMAIN_GTT;
+		place.fpfn = min_offset >> PAGE_SHIFT;
+		place.lpfn = max_offset >> PAGE_SHIFT;
+		place.flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_TT;
+		break;
+	case CGS_GPU_MEM_TYPE__GART_WRITECOMBINE:
+		flags = AMDGPU_GEM_CREATE_CPU_GTT_USWC;
+		domain = AMDGPU_GEM_DOMAIN_GTT;
+		place.fpfn = min_offset >> PAGE_SHIFT;
+		place.lpfn = max_offset >> PAGE_SHIFT;
+		place.flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_TT |
+			TTM_PL_FLAG_UNCACHED;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+
+	*handle = 0;
+
+	placement.placement = &place;
+	placement.num_placement = 1;
+	placement.busy_placement = &place;
+	placement.num_busy_placement = 1;
+
+	ret = amdgpu_bo_create_restricted(adev, size, PAGE_SIZE,
+					  true, domain, flags,
+					  NULL, &placement, &obj);
+	if (ret) {
+		DRM_ERROR("(%d) bo create failed\n", ret);
+		return ret;
+	}
+	*handle = (cgs_handle_t)obj;
+
+	return ret;
+}
+
+static int amdgpu_cgs_import_gpu_mem(void *cgs_device, int dmabuf_fd,
+				     cgs_handle_t *handle)
+{
+	CGS_FUNC_ADEV;
+	int r;
+	uint32_t dma_handle;
+	struct drm_gem_object *obj;
+	struct amdgpu_bo *bo;
+	struct drm_device *dev = adev->ddev;
+	struct drm_file *file_priv = NULL, *priv;
+
+	mutex_lock(&dev->struct_mutex);
+	list_for_each_entry(priv, &dev->filelist, lhead) {
+		rcu_read_lock();
+		if (priv->pid == get_pid(task_pid(current)))
+			file_priv = priv;
+		rcu_read_unlock();
+		if (file_priv)
+			break;
+	}
+	mutex_unlock(&dev->struct_mutex);
+	r = dev->driver->prime_fd_to_handle(dev,
+					    file_priv, dmabuf_fd,
+					    &dma_handle);
+	spin_lock(&file_priv->table_lock);
+
+	/* Check if we currently have a reference on the object */
+	obj = idr_find(&file_priv->object_idr, dma_handle);
+	if (obj == NULL) {
+		spin_unlock(&file_priv->table_lock);
+		return -EINVAL;
+	}
+	spin_unlock(&file_priv->table_lock);
+	bo = gem_to_amdgpu_bo(obj);
+	*handle = (cgs_handle_t)bo;
+	return 0;
+}
+
+static int amdgpu_cgs_free_gpu_mem(void *cgs_device, cgs_handle_t handle)
+{
+	struct amdgpu_bo *obj = (struct amdgpu_bo *)handle;
+
+	if (obj) {
+		int r = amdgpu_bo_reserve(obj, false);
+		if (likely(r == 0)) {
+			amdgpu_bo_kunmap(obj);
+			amdgpu_bo_unpin(obj);
+			amdgpu_bo_unreserve(obj);
+		}
+		amdgpu_bo_unref(&obj);
+
+	}
+	return 0;
+}
+
+static int amdgpu_cgs_gmap_gpu_mem(void *cgs_device, cgs_handle_t handle,
+				   uint64_t *mcaddr)
+{
+	int r;
+	u64 min_offset, max_offset;
+	struct amdgpu_bo *obj = (struct amdgpu_bo *)handle;
+
+	WARN_ON_ONCE(obj->placement.num_placement > 1);
+
+	min_offset = obj->placements[0].fpfn << PAGE_SHIFT;
+	max_offset = obj->placements[0].lpfn << PAGE_SHIFT;
+
+	r = amdgpu_bo_reserve(obj, false);
+	if (unlikely(r != 0))
+		return r;
+	r = amdgpu_bo_pin_restricted(obj, AMDGPU_GEM_DOMAIN_GTT,
+				     min_offset, max_offset, mcaddr);
+	amdgpu_bo_unreserve(obj);
+	return r;
+}
+
+static int amdgpu_cgs_gunmap_gpu_mem(void *cgs_device, cgs_handle_t handle)
+{
+	int r;
+	struct amdgpu_bo *obj = (struct amdgpu_bo *)handle;
+	r = amdgpu_bo_reserve(obj, false);
+	if (unlikely(r != 0))
+		return r;
+	r = amdgpu_bo_unpin(obj);
+	amdgpu_bo_unreserve(obj);
+	return r;
+}
+
+static int amdgpu_cgs_kmap_gpu_mem(void *cgs_device, cgs_handle_t handle,
+				   void **map)
+{
+	int r;
+	struct amdgpu_bo *obj = (struct amdgpu_bo *)handle;
+	r = amdgpu_bo_reserve(obj, false);
+	if (unlikely(r != 0))
+		return r;
+	r = amdgpu_bo_kmap(obj, map);
+	amdgpu_bo_unreserve(obj);
+	return r;
+}
+
+static int amdgpu_cgs_kunmap_gpu_mem(void *cgs_device, cgs_handle_t handle)
+{
+	int r;
+	struct amdgpu_bo *obj = (struct amdgpu_bo *)handle;
+	r = amdgpu_bo_reserve(obj, false);
+	if (unlikely(r != 0))
+		return r;
+	amdgpu_bo_kunmap(obj);
+	amdgpu_bo_unreserve(obj);
+	return r;
+}
+
+static uint32_t amdgpu_cgs_read_register(void *cgs_device, unsigned offset)
+{
+	CGS_FUNC_ADEV;
+	return RREG32(offset);
+}
+
+static void amdgpu_cgs_write_register(void *cgs_device, unsigned offset,
+				      uint32_t value)
+{
+	CGS_FUNC_ADEV;
+	WREG32(offset, value);
+}
+
+static uint32_t amdgpu_cgs_read_ind_register(void *cgs_device,
+					     enum cgs_ind_reg space,
+					     unsigned index)
+{
+	CGS_FUNC_ADEV;
+	switch (space) {
+	case CGS_IND_REG__MMIO:
+		return RREG32_IDX(index);
+	case CGS_IND_REG__PCIE:
+		return RREG32_PCIE(index);
+	case CGS_IND_REG__SMC:
+		return RREG32_SMC(index);
+	case CGS_IND_REG__UVD_CTX:
+		return RREG32_UVD_CTX(index);
+	case CGS_IND_REG__DIDT:
+		return RREG32_DIDT(index);
+	case CGS_IND_REG__AUDIO_ENDPT:
+		DRM_ERROR("audio endpt register access not implemented.\n");
+		return 0;
+	}
+	WARN(1, "Invalid indirect register space");
+	return 0;
+}
+
+static void amdgpu_cgs_write_ind_register(void *cgs_device,
+					  enum cgs_ind_reg space,
+					  unsigned index, uint32_t value)
+{
+	CGS_FUNC_ADEV;
+	switch (space) {
+	case CGS_IND_REG__MMIO:
+		return WREG32_IDX(index, value);
+	case CGS_IND_REG__PCIE:
+		return WREG32_PCIE(index, value);
+	case CGS_IND_REG__SMC:
+		return WREG32_SMC(index, value);
+	case CGS_IND_REG__UVD_CTX:
+		return WREG32_UVD_CTX(index, value);
+	case CGS_IND_REG__DIDT:
+		return WREG32_DIDT(index, value);
+	case CGS_IND_REG__AUDIO_ENDPT:
+		DRM_ERROR("audio endpt register access not implemented.\n");
+		return;
+	}
+	WARN(1, "Invalid indirect register space");
+}
+
+static uint8_t amdgpu_cgs_read_pci_config_byte(void *cgs_device, unsigned addr)
+{
+	CGS_FUNC_ADEV;
+	uint8_t val;
+	int ret = pci_read_config_byte(adev->pdev, addr, &val);
+	if (WARN(ret, "pci_read_config_byte error"))
+		return 0;
+	return val;
+}
+
+static uint16_t amdgpu_cgs_read_pci_config_word(void *cgs_device, unsigned addr)
+{
+	CGS_FUNC_ADEV;
+	uint16_t val;
+	int ret = pci_read_config_word(adev->pdev, addr, &val);
+	if (WARN(ret, "pci_read_config_word error"))
+		return 0;
+	return val;
+}
+
+static uint32_t amdgpu_cgs_read_pci_config_dword(void *cgs_device,
+						 unsigned addr)
+{
+	CGS_FUNC_ADEV;
+	uint32_t val;
+	int ret = pci_read_config_dword(adev->pdev, addr, &val);
+	if (WARN(ret, "pci_read_config_dword error"))
+		return 0;
+	return val;
+}
+
+static void amdgpu_cgs_write_pci_config_byte(void *cgs_device, unsigned addr,
+					     uint8_t value)
+{
+	CGS_FUNC_ADEV;
+	int ret = pci_write_config_byte(adev->pdev, addr, value);
+	WARN(ret, "pci_write_config_byte error");
+}
+
+static void amdgpu_cgs_write_pci_config_word(void *cgs_device, unsigned addr,
+					     uint16_t value)
+{
+	CGS_FUNC_ADEV;
+	int ret = pci_write_config_word(adev->pdev, addr, value);
+	WARN(ret, "pci_write_config_word error");
+}
+
+static void amdgpu_cgs_write_pci_config_dword(void *cgs_device, unsigned addr,
+					      uint32_t value)
+{
+	CGS_FUNC_ADEV;
+	int ret = pci_write_config_dword(adev->pdev, addr, value);
+	WARN(ret, "pci_write_config_dword error");
+}
+
+static const void *amdgpu_cgs_atom_get_data_table(void *cgs_device,
+						  unsigned table, uint16_t *size,
+						  uint8_t *frev, uint8_t *crev)
+{
+	CGS_FUNC_ADEV;
+	uint16_t data_start;
+
+	if (amdgpu_atom_parse_data_header(
+		    adev->mode_info.atom_context, table, size,
+		    frev, crev, &data_start))
+		return (uint8_t*)adev->mode_info.atom_context->bios +
+			data_start;
+
+	return NULL;
+}
+
+static int amdgpu_cgs_atom_get_cmd_table_revs(void *cgs_device, unsigned table,
+					      uint8_t *frev, uint8_t *crev)
+{
+	CGS_FUNC_ADEV;
+
+	if (amdgpu_atom_parse_cmd_header(
+		    adev->mode_info.atom_context, table,
+		    frev, crev))
+		return 0;
+
+	return -EINVAL;
+}
+
+static int amdgpu_cgs_atom_exec_cmd_table(void *cgs_device, unsigned table,
+					  void *args)
+{
+	CGS_FUNC_ADEV;
+
+	return amdgpu_atom_execute_table(
+		adev->mode_info.atom_context, table, args);
+}
+
+static int amdgpu_cgs_create_pm_request(void *cgs_device, cgs_handle_t *request)
+{
+	/* TODO */
+	return 0;
+}
+
+static int amdgpu_cgs_destroy_pm_request(void *cgs_device, cgs_handle_t request)
+{
+	/* TODO */
+	return 0;
+}
+
+static int amdgpu_cgs_set_pm_request(void *cgs_device, cgs_handle_t request,
+				     int active)
+{
+	/* TODO */
+	return 0;
+}
+
+static int amdgpu_cgs_pm_request_clock(void *cgs_device, cgs_handle_t request,
+				       enum cgs_clock clock, unsigned freq)
+{
+	/* TODO */
+	return 0;
+}
+
+static int amdgpu_cgs_pm_request_engine(void *cgs_device, cgs_handle_t request,
+					enum cgs_engine engine, int powered)
+{
+	/* TODO */
+	return 0;
+}
+
+
+
+static int amdgpu_cgs_pm_query_clock_limits(void *cgs_device,
+					    enum cgs_clock clock,
+					    struct cgs_clock_limits *limits)
+{
+	/* TODO */
+	return 0;
+}
+
+static int amdgpu_cgs_set_camera_voltages(void *cgs_device, uint32_t mask,
+					  const uint32_t *voltages)
+{
+	DRM_ERROR("not implemented");
+	return -EPERM;
+}
+
+struct cgs_irq_params {
+	unsigned src_id;
+	cgs_irq_source_set_func_t set;
+	cgs_irq_handler_func_t handler;
+	void *private_data;
+};
+
+static int cgs_set_irq_state(struct amdgpu_device *adev,
+			     struct amdgpu_irq_src *src,
+			     unsigned type,
+			     enum amdgpu_interrupt_state state)
+{
+	struct cgs_irq_params *irq_params =
+		(struct cgs_irq_params *)src->data;
+	if (!irq_params)
+		return -EINVAL;
+	if (!irq_params->set)
+		return -EINVAL;
+	return irq_params->set(irq_params->private_data,
+			       irq_params->src_id,
+			       type,
+			       (int)state);
+}
+
+static int cgs_process_irq(struct amdgpu_device *adev,
+			   struct amdgpu_irq_src *source,
+			   struct amdgpu_iv_entry *entry)
+{
+	struct cgs_irq_params *irq_params =
+		(struct cgs_irq_params *)source->data;
+	if (!irq_params)
+		return -EINVAL;
+	if (!irq_params->handler)
+		return -EINVAL;
+	return irq_params->handler(irq_params->private_data,
+				   irq_params->src_id,
+				   entry->iv_entry);
+}
+
+static const struct amdgpu_irq_src_funcs cgs_irq_funcs = {
+	.set = cgs_set_irq_state,
+	.process = cgs_process_irq,
+};
+
+static int amdgpu_cgs_add_irq_source(void *cgs_device, unsigned src_id,
+				     unsigned num_types,
+				     cgs_irq_source_set_func_t set,
+				     cgs_irq_handler_func_t handler,
+				     void *private_data)
+{
+	CGS_FUNC_ADEV;
+	int ret = 0;
+	struct cgs_irq_params *irq_params;
+	struct amdgpu_irq_src *source =
+		kzalloc(sizeof(struct amdgpu_irq_src), GFP_KERNEL);
+	if (!source)
+		return -ENOMEM;
+	irq_params =
+		kzalloc(sizeof(struct cgs_irq_params), GFP_KERNEL);
+	if (!irq_params) {
+		kfree(source);
+		return -ENOMEM;
+	}
+	source->num_types = num_types;
+	source->funcs = &cgs_irq_funcs;
+	irq_params->src_id = src_id;
+	irq_params->set = set;
+	irq_params->handler = handler;
+	irq_params->private_data = private_data;
+	source->data = (void *)irq_params;
+	ret = amdgpu_irq_add_id(adev, src_id, source);
+	if (ret) {
+		kfree(irq_params);
+		kfree(source);
+	}
+
+	return ret;
+}
+
+static int amdgpu_cgs_irq_get(void *cgs_device, unsigned src_id, unsigned type)
+{
+	CGS_FUNC_ADEV;
+	return amdgpu_irq_get(adev, adev->irq.sources[src_id], type);
+}
+
+static int amdgpu_cgs_irq_put(void *cgs_device, unsigned src_id, unsigned type)
+{
+	CGS_FUNC_ADEV;
+	return amdgpu_irq_put(adev, adev->irq.sources[src_id], type);
+}
+
+int amdgpu_cgs_set_clockgating_state(void *cgs_device,
+				  enum amd_ip_block_type block_type,
+				  enum amd_clockgating_state state)
+{
+	CGS_FUNC_ADEV;
+	int i, r = -1;
+
+	for (i = 0; i < adev->num_ip_blocks; i++) {
+		if (!adev->ip_block_status[i].valid)
+			continue;
+
+		if (adev->ip_blocks[i].type == block_type) {
+			r = adev->ip_blocks[i].funcs->set_clockgating_state(
+								(void *)adev,
+									state);
+			break;
+		}
+	}
+	return r;
+}
+
+int amdgpu_cgs_set_powergating_state(void *cgs_device,
+				  enum amd_ip_block_type block_type,
+				  enum amd_powergating_state state)
+{
+	CGS_FUNC_ADEV;
+	int i, r = -1;
+
+	for (i = 0; i < adev->num_ip_blocks; i++) {
+		if (!adev->ip_block_status[i].valid)
+			continue;
+
+		if (adev->ip_blocks[i].type == block_type) {
+			r = adev->ip_blocks[i].funcs->set_powergating_state(
+								(void *)adev,
+									state);
+			break;
+		}
+	}
+	return r;
+}
+
+
+static uint32_t fw_type_convert(void *cgs_device, uint32_t fw_type)
+{
+	CGS_FUNC_ADEV;
+	enum AMDGPU_UCODE_ID result = AMDGPU_UCODE_ID_MAXIMUM;
+
+	switch (fw_type) {
+	case CGS_UCODE_ID_SDMA0:
+		result = AMDGPU_UCODE_ID_SDMA0;
+		break;
+	case CGS_UCODE_ID_SDMA1:
+		result = AMDGPU_UCODE_ID_SDMA1;
+		break;
+	case CGS_UCODE_ID_CP_CE:
+		result = AMDGPU_UCODE_ID_CP_CE;
+		break;
+	case CGS_UCODE_ID_CP_PFP:
+		result = AMDGPU_UCODE_ID_CP_PFP;
+		break;
+	case CGS_UCODE_ID_CP_ME:
+		result = AMDGPU_UCODE_ID_CP_ME;
+		break;
+	case CGS_UCODE_ID_CP_MEC:
+	case CGS_UCODE_ID_CP_MEC_JT1:
+		result = AMDGPU_UCODE_ID_CP_MEC1;
+		break;
+	case CGS_UCODE_ID_CP_MEC_JT2:
+		if (adev->asic_type == CHIP_TONGA)
+			result = AMDGPU_UCODE_ID_CP_MEC2;
+		else if (adev->asic_type == CHIP_CARRIZO)
+			result = AMDGPU_UCODE_ID_CP_MEC1;
+		break;
+	case CGS_UCODE_ID_RLC_G:
+		result = AMDGPU_UCODE_ID_RLC_G;
+		break;
+	default:
+		DRM_ERROR("Firmware type not supported\n");
+	}
+	return result;
+}
+
+static int amdgpu_cgs_get_firmware_info(void *cgs_device,
+					enum cgs_ucode_id type,
+					struct cgs_firmware_info *info)
+{
+	CGS_FUNC_ADEV;
+
+	if (CGS_UCODE_ID_SMU != type) {
+		uint64_t gpu_addr;
+		uint32_t data_size;
+		const struct gfx_firmware_header_v1_0 *header;
+		enum AMDGPU_UCODE_ID id;
+		struct amdgpu_firmware_info *ucode;
+
+		id = fw_type_convert(cgs_device, type);
+		ucode = &adev->firmware.ucode[id];
+		if (ucode->fw == NULL)
+			return -EINVAL;
+
+		gpu_addr  = ucode->mc_addr;
+		header = (const struct gfx_firmware_header_v1_0 *)ucode->fw->data;
+		data_size = le32_to_cpu(header->header.ucode_size_bytes);
+
+		if ((type == CGS_UCODE_ID_CP_MEC_JT1) ||
+		    (type == CGS_UCODE_ID_CP_MEC_JT2)) {
+			gpu_addr += le32_to_cpu(header->jt_offset) << 2;
+			data_size = le32_to_cpu(header->jt_size) << 2;
+		}
+		info->mc_addr = gpu_addr;
+		info->image_size = data_size;
+		info->version = (uint16_t)le32_to_cpu(header->header.ucode_version);
+		info->feature_version = (uint16_t)le32_to_cpu(header->ucode_feature_version);
+	} else {
+		char fw_name[30] = {0};
+		int err = 0;
+		uint32_t ucode_size;
+		uint32_t ucode_start_address;
+		const uint8_t *src;
+		const struct smc_firmware_header_v1_0 *hdr;
+
+		switch (adev->asic_type) {
+		case CHIP_TONGA:
+			strcpy(fw_name, "amdgpu/tonga_smc.bin");
+			break;
+		default:
+			DRM_ERROR("SMC firmware not supported\n");
+			return -EINVAL;
+		}
+
+		err = request_firmware(&adev->pm.fw, fw_name, adev->dev);
+		if (err) {
+			DRM_ERROR("Failed to request firmware\n");
+			return err;
+		}
+
+		err = amdgpu_ucode_validate(adev->pm.fw);
+		if (err) {
+			DRM_ERROR("Failed to load firmware \"%s\"", fw_name);
+			release_firmware(adev->pm.fw);
+			adev->pm.fw = NULL;
+			return err;
+		}
+
+		hdr = (const struct smc_firmware_header_v1_0 *)	adev->pm.fw->data;
+		adev->pm.fw_version = le32_to_cpu(hdr->header.ucode_version);
+		ucode_size = le32_to_cpu(hdr->header.ucode_size_bytes);
+		ucode_start_address = le32_to_cpu(hdr->ucode_start_addr);
+		src = (const uint8_t *)(adev->pm.fw->data +
+		       le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+
+		info->version = adev->pm.fw_version;
+		info->image_size = ucode_size;
+		info->kptr = (void *)src;
+	}
+	return 0;
+}
+
+static const struct cgs_ops amdgpu_cgs_ops = {
+	amdgpu_cgs_gpu_mem_info,
+	amdgpu_cgs_gmap_kmem,
+	amdgpu_cgs_gunmap_kmem,
+	amdgpu_cgs_alloc_gpu_mem,
+	amdgpu_cgs_free_gpu_mem,
+	amdgpu_cgs_gmap_gpu_mem,
+	amdgpu_cgs_gunmap_gpu_mem,
+	amdgpu_cgs_kmap_gpu_mem,
+	amdgpu_cgs_kunmap_gpu_mem,
+	amdgpu_cgs_read_register,
+	amdgpu_cgs_write_register,
+	amdgpu_cgs_read_ind_register,
+	amdgpu_cgs_write_ind_register,
+	amdgpu_cgs_read_pci_config_byte,
+	amdgpu_cgs_read_pci_config_word,
+	amdgpu_cgs_read_pci_config_dword,
+	amdgpu_cgs_write_pci_config_byte,
+	amdgpu_cgs_write_pci_config_word,
+	amdgpu_cgs_write_pci_config_dword,
+	amdgpu_cgs_atom_get_data_table,
+	amdgpu_cgs_atom_get_cmd_table_revs,
+	amdgpu_cgs_atom_exec_cmd_table,
+	amdgpu_cgs_create_pm_request,
+	amdgpu_cgs_destroy_pm_request,
+	amdgpu_cgs_set_pm_request,
+	amdgpu_cgs_pm_request_clock,
+	amdgpu_cgs_pm_request_engine,
+	amdgpu_cgs_pm_query_clock_limits,
+	amdgpu_cgs_set_camera_voltages,
+	amdgpu_cgs_get_firmware_info,
+	amdgpu_cgs_set_powergating_state,
+	amdgpu_cgs_set_clockgating_state
+};
+
+static const struct cgs_os_ops amdgpu_cgs_os_ops = {
+	amdgpu_cgs_import_gpu_mem,
+	amdgpu_cgs_add_irq_source,
+	amdgpu_cgs_irq_get,
+	amdgpu_cgs_irq_put
+};
+
+void *amdgpu_cgs_create_device(struct amdgpu_device *adev)
+{
+	struct amdgpu_cgs_device *cgs_device =
+		kmalloc(sizeof(*cgs_device), GFP_KERNEL);
+
+	if (!cgs_device) {
+		DRM_ERROR("Couldn't allocate CGS device structure\n");
+		return NULL;
+	}
+
+	cgs_device->base.ops = &amdgpu_cgs_ops;
+	cgs_device->base.os_ops = &amdgpu_cgs_os_ops;
+	cgs_device->adev = adev;
+
+	return cgs_device;
+}
+
+void amdgpu_cgs_destroy_device(void *cgs_device)
+{
+	kfree(cgs_device);
+}
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@ -126,12 +126,54 @@ int amdgpu_cs_get_ring(struct amdgpu_device *adev, u32 ip_type,
 	return 0;
 }

+static void amdgpu_job_work_func(struct work_struct *work)
+{
+	struct amdgpu_cs_parser *sched_job =
+		container_of(work, struct amdgpu_cs_parser,
+			     job_work);
+	mutex_lock(&sched_job->job_lock);
+	if (sched_job->free_job)
+		sched_job->free_job(sched_job);
+	mutex_unlock(&sched_job->job_lock);
+	/* after processing job, free memory */
+	fence_put(&sched_job->s_fence->base);
+	kfree(sched_job);
+}
+struct amdgpu_cs_parser *amdgpu_cs_parser_create(struct amdgpu_device *adev,
+                                               struct drm_file *filp,
+                                               struct amdgpu_ctx *ctx,
+                                               struct amdgpu_ib *ibs,
+                                               uint32_t num_ibs)
+{
+	struct amdgpu_cs_parser *parser;
+	int i;
+
+	parser = kzalloc(sizeof(struct amdgpu_cs_parser), GFP_KERNEL);
+	if (!parser)
+		return NULL;
+
+	parser->adev = adev;
+	parser->filp = filp;
+	parser->ctx = ctx;
+	parser->ibs = ibs;
+	parser->num_ibs = num_ibs;
+	if (amdgpu_enable_scheduler) {
+		mutex_init(&parser->job_lock);
+		INIT_WORK(&parser->job_work, amdgpu_job_work_func);
+	}
+	for (i = 0; i < num_ibs; i++)
+		ibs[i].ctx = ctx;
+
+	return parser;
+}
+
 int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 {
 	union drm_amdgpu_cs *cs = data;
 	uint64_t *chunk_array_user;
 	uint64_t *chunk_array = NULL;
 	struct amdgpu_fpriv *fpriv = p->filp->driver_priv;
+	struct amdgpu_bo_list *bo_list = NULL;
 	unsigned size, i;
 	int r = 0;

@ -143,17 +185,30 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 		r = -EINVAL;
 		goto out;
 	}
-	p->bo_list = amdgpu_bo_list_get(fpriv, cs->in.bo_list_handle);
+	bo_list = amdgpu_bo_list_get(fpriv, cs->in.bo_list_handle);
+	if (!amdgpu_enable_scheduler)
+		p->bo_list = bo_list;
+	else {
+		if (bo_list && !bo_list->has_userptr) {
+			p->bo_list = amdgpu_bo_list_clone(bo_list);
+			amdgpu_bo_list_put(bo_list);
+			if (!p->bo_list)
+				return -ENOMEM;
+		} else if (bo_list && bo_list->has_userptr)
+			p->bo_list = bo_list;
+		else
+			p->bo_list = NULL;
+	}

 	/* get chunks */
 	INIT_LIST_HEAD(&p->validated);
-	chunk_array = kcalloc(cs->in.num_chunks, sizeof(uint64_t), GFP_KERNEL);
+	chunk_array = kmalloc_array(cs->in.num_chunks, sizeof(uint64_t), GFP_KERNEL);
 	if (chunk_array == NULL) {
 		r = -ENOMEM;
 		goto out;
 	}

-	chunk_array_user = (uint64_t *)(unsigned long)(cs->in.chunks);
+	chunk_array_user = (uint64_t __user *)(cs->in.chunks);
 	if (copy_from_user(chunk_array, chunk_array_user,
 			   sizeof(uint64_t)*cs->in.num_chunks)) {
 		r = -EFAULT;
@ -161,7 +216,7 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 	}

 	p->nchunks = cs->in.num_chunks;
-	p->chunks = kcalloc(p->nchunks, sizeof(struct amdgpu_cs_chunk),
+	p->chunks = kmalloc_array(p->nchunks, sizeof(struct amdgpu_cs_chunk),
 			    GFP_KERNEL);
 	if (p->chunks == NULL) {
 		r = -ENOMEM;
@ -173,7 +228,7 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 		struct drm_amdgpu_cs_chunk user_chunk;
 		uint32_t __user *cdata;

-		chunk_ptr = (void __user *)(unsigned long)chunk_array[i];
+		chunk_ptr = (void __user *)chunk_array[i];
 		if (copy_from_user(&user_chunk, chunk_ptr,
 				       sizeof(struct drm_amdgpu_cs_chunk))) {
 			r = -EFAULT;
@ -183,7 +238,7 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 		p->chunks[i].length_dw = user_chunk.length_dw;

 		size = p->chunks[i].length_dw;
-		cdata = (void __user *)(unsigned long)user_chunk.chunk_data;
+		cdata = (void __user *)user_chunk.chunk_data;
 		p->chunks[i].user_ptr = cdata;

 		p->chunks[i].kdata = drm_malloc_ab(size, sizeof(uint32_t));
@ -235,11 +290,10 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 		}
 	}

-	p->ibs = kcalloc(p->num_ibs, sizeof(struct amdgpu_ib), GFP_KERNEL);
-	if (!p->ibs) {
+
+	p->ibs = kmalloc_array(p->num_ibs, sizeof(struct amdgpu_ib), GFP_KERNEL);
+	if (!p->ibs)
 		r = -ENOMEM;
-		goto out;
-	}

 out:
 	kfree(chunk_array);
@ -415,18 +469,8 @@ static int cmp_size_smaller_first(void *priv, struct list_head *a,
 	return (int)la->robj->tbo.num_pages - (int)lb->robj->tbo.num_pages;
 }

-/**
- * cs_parser_fini() - clean parser states
- * @parser:	parser structure holding parsing context.
- * @error:	error number
- *
- * If error is set than unvalidate buffer, otherwise just free memory
- * used by parsing context.
- **/
-static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, bool backoff)
+static void amdgpu_cs_parser_fini_early(struct amdgpu_cs_parser *parser, int error, bool backoff)
 {
-	unsigned i;
-
 	if (!error) {
 		/* Sort the buffer list from the smallest to largest buffer,
 		 * which affects the order of buffers in the LRU list.
@ -447,11 +491,19 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, bo
 		ttm_eu_backoff_reservation(&parser->ticket,
 					   &parser->validated);
 	}
+}

+static void amdgpu_cs_parser_fini_late(struct amdgpu_cs_parser *parser)
+{
+	unsigned i;
 	if (parser->ctx)
 		amdgpu_ctx_put(parser->ctx);
-	if (parser->bo_list)
-		amdgpu_bo_list_put(parser->bo_list);
+	if (parser->bo_list) {
+		if (amdgpu_enable_scheduler && !parser->bo_list->has_userptr)
+			amdgpu_bo_list_free(parser->bo_list);
+		else
+			amdgpu_bo_list_put(parser->bo_list);
+	}
 	drm_free_large(parser->vm_bos);
 	for (i = 0; i < parser->nchunks; i++)
 		drm_free_large(parser->chunks[i].kdata);
@ -462,6 +514,29 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, bo
 	kfree(parser->ibs);
 	if (parser->uf.bo)
 		drm_gem_object_unreference_unlocked(&parser->uf.bo->gem_base);
+
+	if (!amdgpu_enable_scheduler)
+		kfree(parser);
+}
+
+/**
+ * cs_parser_fini() - clean parser states
+ * @parser:	parser structure holding parsing context.
+ * @error:	error number
+ *
+ * If error is set than unvalidate buffer, otherwise just free memory
+ * used by parsing context.
+ **/
+static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, bool backoff)
+{
+       amdgpu_cs_parser_fini_early(parser, error, backoff);
+       amdgpu_cs_parser_fini_late(parser);
+}
+
+static int amdgpu_cs_parser_free_job(struct amdgpu_cs_parser *sched_job)
+{
+       amdgpu_cs_parser_fini_late(sched_job);
+       return 0;
 }

 static int amdgpu_bo_vm_update_pte(struct amdgpu_cs_parser *p,
@ -476,12 +551,18 @@ static int amdgpu_bo_vm_update_pte(struct amdgpu_cs_parser *p,
 	if (r)
 		return r;

+	r = amdgpu_sync_fence(adev, &p->ibs[0].sync, vm->page_directory_fence);
+	if (r)
+		return r;
+
 	r = amdgpu_vm_clear_freed(adev, vm);
 	if (r)
 		return r;

 	if (p->bo_list) {
 		for (i = 0; i < p->bo_list->num_entries; i++) {
+			struct fence *f;
+
 			/* ignore duplicates */
 			bo = p->bo_list->array[i].robj;
 			if (!bo)
@ -495,7 +576,10 @@ static int amdgpu_bo_vm_update_pte(struct amdgpu_cs_parser *p,
 			if (r)
 				return r;

-			amdgpu_sync_fence(&p->ibs[0].sync, bo_va->last_pt_update);
+			f = bo_va->last_pt_update;
+			r = amdgpu_sync_fence(adev, &p->ibs[0].sync, f);
+			if (r)
+				return r;
 		}
 	}

@ -529,9 +613,9 @@ static int amdgpu_cs_ib_vm_chunk(struct amdgpu_device *adev,
 		goto out;
 	}
 	amdgpu_cs_sync_rings(parser);
-
-	r = amdgpu_ib_schedule(adev, parser->num_ibs, parser->ibs,
-			       parser->filp);
+	if (!amdgpu_enable_scheduler)
+		r = amdgpu_ib_schedule(adev, parser->num_ibs, parser->ibs,
+				       parser->filp);

 out:
 	mutex_unlock(&vm->mutex);
@ -650,7 +734,6 @@ static int amdgpu_cs_ib_fill(struct amdgpu_device *adev,
 			ib->oa_size = amdgpu_bo_size(oa);
 		}
 	}
-
 	/* wrap the last IB with user fence */
 	if (parser->uf.bo) {
 		struct amdgpu_ib *ib = &parser->ibs[parser->num_ibs - 1];
@ -693,9 +776,9 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
 			sizeof(struct drm_amdgpu_cs_chunk_dep);

 		for (j = 0; j < num_deps; ++j) {
-			struct amdgpu_fence *fence;
 			struct amdgpu_ring *ring;
 			struct amdgpu_ctx *ctx;
+			struct fence *fence;

 			r = amdgpu_cs_get_ring(adev, deps[j].ip_type,
 					       deps[j].ip_instance,
@ -707,50 +790,34 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
 			if (ctx == NULL)
 				return -EINVAL;

-			r = amdgpu_fence_recreate(ring, p->filp,
-						  deps[j].handle,
-						  &fence);
-			if (r) {
+			fence = amdgpu_ctx_get_fence(ctx, ring,
+						     deps[j].handle);
+			if (IS_ERR(fence)) {
+				r = PTR_ERR(fence);
 				amdgpu_ctx_put(ctx);
 				return r;
-			}

-			amdgpu_sync_fence(&ib->sync, fence);
-			amdgpu_fence_unref(&fence);
-			amdgpu_ctx_put(ctx);
+			} else if (fence) {
+				r = amdgpu_sync_fence(adev, &ib->sync, fence);
+				fence_put(fence);
+				amdgpu_ctx_put(ctx);
+				if (r)
+					return r;
+			}
 		}
 	}

 	return 0;
 }

-int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
+static int amdgpu_cs_parser_prepare_job(struct amdgpu_cs_parser *sched_job)
 {
-	struct amdgpu_device *adev = dev->dev_private;
-	union drm_amdgpu_cs *cs = data;
-	struct amdgpu_cs_parser parser;
 	int r, i;
+	struct amdgpu_cs_parser *parser = sched_job;
+	struct amdgpu_device *adev = sched_job->adev;
 	bool reserved_buffers = false;

-	down_read(&adev->exclusive_lock);
-	if (!adev->accel_working) {
-		up_read(&adev->exclusive_lock);
-		return -EBUSY;
-	}
-	/* initialize parser */
-	memset(&parser, 0, sizeof(struct amdgpu_cs_parser));
-	parser.filp = filp;
-	parser.adev = adev;
-	r = amdgpu_cs_parser_init(&parser, data);
-	if (r) {
-		DRM_ERROR("Failed to initialize parser !\n");
-		amdgpu_cs_parser_fini(&parser, r, false);
-		up_read(&adev->exclusive_lock);
-		r = amdgpu_cs_handle_lockup(adev, r);
-		return r;
-	}
-
-	r = amdgpu_cs_parser_relocs(&parser);
+	r = amdgpu_cs_parser_relocs(parser);
 	if (r) {
 		if (r != -ERESTARTSYS) {
 			if (r == -ENOMEM)
@ -762,30 +829,114 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)

 	if (!r) {
 		reserved_buffers = true;
-		r = amdgpu_cs_ib_fill(adev, &parser);
+		r = amdgpu_cs_ib_fill(adev, parser);
+	}
+	if (!r) {
+		r = amdgpu_cs_dependencies(adev, parser);
+		if (r)
+			DRM_ERROR("Failed in the dependencies handling %d!\n", r);
+	}
+	if (r) {
+		amdgpu_cs_parser_fini(parser, r, reserved_buffers);
+		return r;
 	}

-	if (!r)
-		r = amdgpu_cs_dependencies(adev, &parser);
+	for (i = 0; i < parser->num_ibs; i++)
+		trace_amdgpu_cs(parser, i);

+	r = amdgpu_cs_ib_vm_chunk(adev, parser);
+	return r;
+}
+
+static struct amdgpu_ring *amdgpu_cs_parser_get_ring(
+	struct amdgpu_device *adev,
+	struct amdgpu_cs_parser *parser)
+{
+	int i, r;
+
+	struct amdgpu_cs_chunk *chunk;
+	struct drm_amdgpu_cs_chunk_ib *chunk_ib;
+	struct amdgpu_ring *ring;
+	for (i = 0; i < parser->nchunks; i++) {
+		chunk = &parser->chunks[i];
+		chunk_ib = (struct drm_amdgpu_cs_chunk_ib *)chunk->kdata;
+
+		if (chunk->chunk_id != AMDGPU_CHUNK_ID_IB)
+			continue;
+
+		r = amdgpu_cs_get_ring(adev, chunk_ib->ip_type,
+				       chunk_ib->ip_instance, chunk_ib->ring,
+				       &ring);
+		if (r)
+			return NULL;
+		break;
+	}
+	return ring;
+}
+
+int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
+{
+	struct amdgpu_device *adev = dev->dev_private;
+	union drm_amdgpu_cs *cs = data;
+	struct amdgpu_cs_parser *parser;
+	int r;
+
+	down_read(&adev->exclusive_lock);
+	if (!adev->accel_working) {
+		up_read(&adev->exclusive_lock);
+		return -EBUSY;
+	}
+
+	parser = amdgpu_cs_parser_create(adev, filp, NULL, NULL, 0);
+	if (!parser)
+		return -ENOMEM;
+	r = amdgpu_cs_parser_init(parser, data);
 	if (r) {
-		amdgpu_cs_parser_fini(&parser, r, reserved_buffers);
+		DRM_ERROR("Failed to initialize parser !\n");
+		amdgpu_cs_parser_fini(parser, r, false);
 		up_read(&adev->exclusive_lock);
 		r = amdgpu_cs_handle_lockup(adev, r);
 		return r;
 	}

-	for (i = 0; i < parser.num_ibs; i++)
-		trace_amdgpu_cs(&parser, i);
+	if (amdgpu_enable_scheduler && parser->num_ibs) {
+		struct amdgpu_ring * ring =
+			amdgpu_cs_parser_get_ring(adev, parser);
+		r = amdgpu_cs_parser_prepare_job(parser);
+		if (r)
+			goto out;
+		parser->ring = ring;
+		parser->free_job = amdgpu_cs_parser_free_job;
+		mutex_lock(&parser->job_lock);
+		r = amd_sched_push_job(ring->scheduler,
+				       &parser->ctx->rings[ring->idx].entity,
+				       parser,
+				       &parser->s_fence);
+		if (r) {
+			mutex_unlock(&parser->job_lock);
+			goto out;
+		}
+		parser->ibs[parser->num_ibs - 1].sequence =
+			amdgpu_ctx_add_fence(parser->ctx, ring,
+					     &parser->s_fence->base,
+					     parser->s_fence->v_seq);
+		cs->out.handle = parser->s_fence->v_seq;
+		list_sort(NULL, &parser->validated, cmp_size_smaller_first);
+		ttm_eu_fence_buffer_objects(&parser->ticket,
+				&parser->validated,
+				&parser->s_fence->base);

-	r = amdgpu_cs_ib_vm_chunk(adev, &parser);
-	if (r) {
-		goto out;
+		mutex_unlock(&parser->job_lock);
+		up_read(&adev->exclusive_lock);
+		return 0;
 	}
+	r = amdgpu_cs_parser_prepare_job(parser);
+	if (r)
+		goto out;

-	cs->out.handle = parser.ibs[parser.num_ibs - 1].fence->seq;
+	cs->out.handle = parser->ibs[parser->num_ibs - 1].sequence;
 out:
-	amdgpu_cs_parser_fini(&parser, r, true);
+	amdgpu_cs_parser_fini(parser, r, true);
 	up_read(&adev->exclusive_lock);
 	r = amdgpu_cs_handle_lockup(adev, r);
 	return r;
@ -806,30 +957,29 @@ int amdgpu_cs_wait_ioctl(struct drm_device *dev, void *data,
 	union drm_amdgpu_wait_cs *wait = data;
 	struct amdgpu_device *adev = dev->dev_private;
 	unsigned long timeout = amdgpu_gem_timeout(wait->in.timeout);
-	struct amdgpu_fence *fence = NULL;
 	struct amdgpu_ring *ring = NULL;
 	struct amdgpu_ctx *ctx;
+	struct fence *fence;
 	long r;

+	r = amdgpu_cs_get_ring(adev, wait->in.ip_type, wait->in.ip_instance,
+			       wait->in.ring, &ring);
+	if (r)
+		return r;
+
 	ctx = amdgpu_ctx_get(filp->driver_priv, wait->in.ctx_id);
 	if (ctx == NULL)
 		return -EINVAL;

-	r = amdgpu_cs_get_ring(adev, wait->in.ip_type, wait->in.ip_instance,
-			       wait->in.ring, &ring);
-	if (r) {
-		amdgpu_ctx_put(ctx);
-		return r;
-	}
+	fence = amdgpu_ctx_get_fence(ctx, ring, wait->in.handle);
+	if (IS_ERR(fence))
+		r = PTR_ERR(fence);
+	else if (fence) {
+		r = fence_wait_timeout(fence, true, timeout);
+		fence_put(fence);
+	} else
+		r = 1;

-	r = amdgpu_fence_recreate(ring, filp, wait->in.handle, &fence);
-	if (r) {
-		amdgpu_ctx_put(ctx);
-		return r;
-	}
-
-	r = fence_wait_timeout(&fence->base, true, timeout);
-	amdgpu_fence_unref(&fence);
 	amdgpu_ctx_put(ctx);
 	if (r < 0)
 		return r;
@ -864,7 +1014,16 @@ amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser,
 		if (!reloc->bo_va)
 			continue;

-		list_for_each_entry(mapping, &reloc->bo_va->mappings, list) {
+		list_for_each_entry(mapping, &reloc->bo_va->valids, list) {
+			if (mapping->it.start > addr ||
+			    addr > mapping->it.last)
+				continue;
+
+			*bo = reloc->bo_va->bo;
+			return mapping;
+		}
+
+		list_for_each_entry(mapping, &reloc->bo_va->invalids, list) {
 			if (mapping->it.start > addr ||
 			    addr > mapping->it.last)
 				continue;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@ -25,54 +25,107 @@
 #include <drm/drmP.h>
 #include "amdgpu.h"

-static void amdgpu_ctx_do_release(struct kref *ref)
+int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel,
+		    struct amdgpu_ctx *ctx)
 {
-	struct amdgpu_ctx *ctx;
-	struct amdgpu_ctx_mgr *mgr;
+	unsigned i, j;
+	int r;

-	ctx = container_of(ref, struct amdgpu_ctx, refcount);
-	mgr = &ctx->fpriv->ctx_mgr;
+	memset(ctx, 0, sizeof(*ctx));
+	ctx->adev = adev;
+	kref_init(&ctx->refcount);
+	spin_lock_init(&ctx->ring_lock);
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i)
+		ctx->rings[i].sequence = 1;

-	idr_remove(&mgr->ctx_handles, ctx->id);
-	kfree(ctx);
+	if (amdgpu_enable_scheduler) {
+		/* create context entity for each ring */
+		for (i = 0; i < adev->num_rings; i++) {
+			struct amd_sched_rq *rq;
+			if (kernel)
+				rq = &adev->rings[i]->scheduler->kernel_rq;
+			else
+				rq = &adev->rings[i]->scheduler->sched_rq;
+			r = amd_sched_entity_init(adev->rings[i]->scheduler,
+						  &ctx->rings[i].entity,
+						  rq, amdgpu_sched_jobs);
+			if (r)
+				break;
+		}
+
+		if (i < adev->num_rings) {
+			for (j = 0; j < i; j++)
+				amd_sched_entity_fini(adev->rings[j]->scheduler,
+						      &ctx->rings[j].entity);
+			kfree(ctx);
+			return r;
+		}
+	}
+	return 0;
 }

-int amdgpu_ctx_alloc(struct amdgpu_device *adev, struct amdgpu_fpriv *fpriv, uint32_t *id, uint32_t flags)
+void amdgpu_ctx_fini(struct amdgpu_ctx *ctx)
+{
+	struct amdgpu_device *adev = ctx->adev;
+	unsigned i, j;
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i)
+		for (j = 0; j < AMDGPU_CTX_MAX_CS_PENDING; ++j)
+			fence_put(ctx->rings[i].fences[j]);
+
+	if (amdgpu_enable_scheduler) {
+		for (i = 0; i < adev->num_rings; i++)
+			amd_sched_entity_fini(adev->rings[i]->scheduler,
+					      &ctx->rings[i].entity);
+	}
+}
+
+static int amdgpu_ctx_alloc(struct amdgpu_device *adev,
+			    struct amdgpu_fpriv *fpriv,
+			    uint32_t *id)
 {
-	int r;
-	struct amdgpu_ctx *ctx;
 	struct amdgpu_ctx_mgr *mgr = &fpriv->ctx_mgr;
+	struct amdgpu_ctx *ctx;
+	int r;

 	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
 		return -ENOMEM;

 	mutex_lock(&mgr->lock);
-	r = idr_alloc(&mgr->ctx_handles, ctx, 0, 0, GFP_KERNEL);
+	r = idr_alloc(&mgr->ctx_handles, ctx, 1, 0, GFP_KERNEL);
 	if (r < 0) {
 		mutex_unlock(&mgr->lock);
 		kfree(ctx);
 		return r;
 	}
 	*id = (uint32_t)r;
-
-	memset(ctx, 0, sizeof(*ctx));
-	ctx->id = *id;
-	ctx->fpriv = fpriv;
-	kref_init(&ctx->refcount);
+	r = amdgpu_ctx_init(adev, false, ctx);
 	mutex_unlock(&mgr->lock);

-	return 0;
+	return r;
 }

-int amdgpu_ctx_free(struct amdgpu_device *adev, struct amdgpu_fpriv *fpriv, uint32_t id)
+static void amdgpu_ctx_do_release(struct kref *ref)
 {
 	struct amdgpu_ctx *ctx;
+
+	ctx = container_of(ref, struct amdgpu_ctx, refcount);
+
+	amdgpu_ctx_fini(ctx);
+
+	kfree(ctx);
+}
+
+static int amdgpu_ctx_free(struct amdgpu_fpriv *fpriv, uint32_t id)
+{
 	struct amdgpu_ctx_mgr *mgr = &fpriv->ctx_mgr;
+	struct amdgpu_ctx *ctx;

 	mutex_lock(&mgr->lock);
 	ctx = idr_find(&mgr->ctx_handles, id);
 	if (ctx) {
+		idr_remove(&mgr->ctx_handles, id);
 		kref_put(&ctx->refcount, amdgpu_ctx_do_release);
 		mutex_unlock(&mgr->lock);
 		return 0;
@ -86,9 +139,13 @@ static int amdgpu_ctx_query(struct amdgpu_device *adev,
 			    union drm_amdgpu_ctx_out *out)
 {
 	struct amdgpu_ctx *ctx;
-	struct amdgpu_ctx_mgr *mgr = &fpriv->ctx_mgr;
+	struct amdgpu_ctx_mgr *mgr;
 	unsigned reset_counter;

+	if (!fpriv)
+		return -EINVAL;
+
+	mgr = &fpriv->ctx_mgr;
 	mutex_lock(&mgr->lock);
 	ctx = idr_find(&mgr->ctx_handles, id);
 	if (!ctx) {
@ -97,8 +154,8 @@ static int amdgpu_ctx_query(struct amdgpu_device *adev,
 	}

 	/* TODO: these two are always zero */
-	out->state.flags = ctx->state.flags;
-	out->state.hangs = ctx->state.hangs;
+	out->state.flags = 0x0;
+	out->state.hangs = 0x0;

 	/* determine if a GPU reset has occured since the last call */
 	reset_counter = atomic_read(&adev->gpu_reset_counter);
@ -113,28 +170,11 @@ static int amdgpu_ctx_query(struct amdgpu_device *adev,
 	return 0;
 }

-void amdgpu_ctx_fini(struct amdgpu_fpriv *fpriv)
-{
-	struct idr *idp;
-	struct amdgpu_ctx *ctx;
-	uint32_t id;
-	struct amdgpu_ctx_mgr *mgr = &fpriv->ctx_mgr;
-	idp = &mgr->ctx_handles;
-
-	idr_for_each_entry(idp,ctx,id) {
-		if (kref_put(&ctx->refcount, amdgpu_ctx_do_release) != 1)
-			DRM_ERROR("ctx (id=%ul) is still alive\n",ctx->id);
-	}
-
-	mutex_destroy(&mgr->lock);
-}
-
 int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
 		     struct drm_file *filp)
 {
 	int r;
 	uint32_t id;
-	uint32_t flags;

 	union drm_amdgpu_ctx *args = data;
 	struct amdgpu_device *adev = dev->dev_private;
@ -142,15 +182,14 @@ int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,

 	r = 0;
 	id = args->in.ctx_id;
-	flags = args->in.flags;

 	switch (args->in.op) {
 		case AMDGPU_CTX_OP_ALLOC_CTX:
-			r = amdgpu_ctx_alloc(adev, fpriv, &id, flags);
+			r = amdgpu_ctx_alloc(adev, fpriv, &id);
 			args->out.alloc.ctx_id = id;
 			break;
 		case AMDGPU_CTX_OP_FREE_CTX:
-			r = amdgpu_ctx_free(adev, fpriv, id);
+			r = amdgpu_ctx_free(fpriv, id);
 			break;
 		case AMDGPU_CTX_OP_QUERY_STATE:
 			r = amdgpu_ctx_query(adev, fpriv, id, &args->out);
@ -165,7 +204,12 @@ int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
 struct amdgpu_ctx *amdgpu_ctx_get(struct amdgpu_fpriv *fpriv, uint32_t id)
 {
 	struct amdgpu_ctx *ctx;
-	struct amdgpu_ctx_mgr *mgr = &fpriv->ctx_mgr;
+	struct amdgpu_ctx_mgr *mgr;
+
+	if (!fpriv)
+		return NULL;
+
+	mgr = &fpriv->ctx_mgr;

 	mutex_lock(&mgr->lock);
 	ctx = idr_find(&mgr->ctx_handles, id);
@ -177,17 +221,96 @@ struct amdgpu_ctx *amdgpu_ctx_get(struct amdgpu_fpriv *fpriv, uint32_t id)

 int amdgpu_ctx_put(struct amdgpu_ctx *ctx)
 {
-	struct amdgpu_fpriv *fpriv;
-	struct amdgpu_ctx_mgr *mgr;
-
 	if (ctx == NULL)
 		return -EINVAL;

-	fpriv = ctx->fpriv;
-	mgr = &fpriv->ctx_mgr;
-	mutex_lock(&mgr->lock);
 	kref_put(&ctx->refcount, amdgpu_ctx_do_release);
-	mutex_unlock(&mgr->lock);
-
 	return 0;
 }
+
+uint64_t amdgpu_ctx_add_fence(struct amdgpu_ctx *ctx, struct amdgpu_ring *ring,
+			      struct fence *fence, uint64_t queued_seq)
+{
+	struct amdgpu_ctx_ring *cring = & ctx->rings[ring->idx];
+	uint64_t seq = 0;
+	unsigned idx = 0;
+	struct fence *other = NULL;
+
+	if (amdgpu_enable_scheduler)
+		seq = queued_seq;
+	else
+		seq = cring->sequence;
+	idx = seq % AMDGPU_CTX_MAX_CS_PENDING;
+	other = cring->fences[idx];
+	if (other) {
+		signed long r;
+		r = fence_wait_timeout(other, false, MAX_SCHEDULE_TIMEOUT);
+		if (r < 0)
+			DRM_ERROR("Error (%ld) waiting for fence!\n", r);
+	}
+
+	fence_get(fence);
+
+	spin_lock(&ctx->ring_lock);
+	cring->fences[idx] = fence;
+	if (!amdgpu_enable_scheduler)
+		cring->sequence++;
+	spin_unlock(&ctx->ring_lock);
+
+	fence_put(other);
+
+	return seq;
+}
+
+struct fence *amdgpu_ctx_get_fence(struct amdgpu_ctx *ctx,
+				   struct amdgpu_ring *ring, uint64_t seq)
+{
+	struct amdgpu_ctx_ring *cring = & ctx->rings[ring->idx];
+	struct fence *fence;
+	uint64_t queued_seq;
+
+	spin_lock(&ctx->ring_lock);
+	if (amdgpu_enable_scheduler)
+		queued_seq = amd_sched_next_queued_seq(&cring->entity);
+	else
+		queued_seq = cring->sequence;
+
+	if (seq >= queued_seq) {
+		spin_unlock(&ctx->ring_lock);
+		return ERR_PTR(-EINVAL);
+	}
+
+
+	if (seq + AMDGPU_CTX_MAX_CS_PENDING < queued_seq) {
+		spin_unlock(&ctx->ring_lock);
+		return NULL;
+	}
+
+	fence = fence_get(cring->fences[seq % AMDGPU_CTX_MAX_CS_PENDING]);
+	spin_unlock(&ctx->ring_lock);
+
+	return fence;
+}
+
+void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr)
+{
+	mutex_init(&mgr->lock);
+	idr_init(&mgr->ctx_handles);
+}
+
+void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr)
+{
+	struct amdgpu_ctx *ctx;
+	struct idr *idp;
+	uint32_t id;
+
+	idp = &mgr->ctx_handles;
+
+	idr_for_each_entry(idp, ctx, id) {
+		if (kref_put(&ctx->refcount, amdgpu_ctx_do_release) != 1)
+			DRM_ERROR("ctx %p is still alive\n", ctx);
+	}
+
+	idr_destroy(&mgr->ctx_handles);
+	mutex_destroy(&mgr->lock);
+}
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@ -55,6 +55,7 @@ static const char *amdgpu_asic_name[] = {
 	"MULLINS",
 	"TOPAZ",
 	"TONGA",
+	"FIJI",
 	"CARRIZO",
 	"LAST",
 };
@ -63,7 +64,7 @@ bool amdgpu_device_is_px(struct drm_device *dev)
 {
 	struct amdgpu_device *adev = dev->dev_private;

-	if (adev->flags & AMDGPU_IS_PX)
+	if (adev->flags & AMD_IS_PX)
 		return true;
 	return false;
 }
@ -1160,6 +1161,7 @@ static int amdgpu_early_init(struct amdgpu_device *adev)
 	switch (adev->asic_type) {
 	case CHIP_TOPAZ:
 	case CHIP_TONGA:
+	case CHIP_FIJI:
 	case CHIP_CARRIZO:
 		if (adev->asic_type == CHIP_CARRIZO)
 			adev->family = AMDGPU_FAMILY_CZ;
@ -1377,7 +1379,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	adev->ddev = ddev;
 	adev->pdev = pdev;
 	adev->flags = flags;
-	adev->asic_type = flags & AMDGPU_ASIC_MASK;
+	adev->asic_type = flags & AMD_ASIC_MASK;
 	adev->is_atom_bios = false;
 	adev->usec_timeout = AMDGPU_MAX_USEC_TIMEOUT;
 	adev->mc.gtt_size = 512 * 1024 * 1024;
@ -1523,6 +1525,11 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 		return r;
 	}

+	r = amdgpu_ctx_init(adev, true, &adev->kernel_ctx);
+	if (r) {
+		dev_err(adev->dev, "failed to create kernel context (%d).\n", r);
+		return r;
+	}
 	r = amdgpu_ib_ring_tests(adev);
 	if (r)
 		DRM_ERROR("ib ring test failed (%d).\n", r);
@ -1584,6 +1591,7 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 	adev->shutdown = true;
 	/* evict vram memory */
 	amdgpu_bo_evict_vram(adev);
+	amdgpu_ctx_fini(&adev->kernel_ctx);
 	amdgpu_ib_pool_fini(adev);
 	amdgpu_fence_driver_fini(adev);
 	amdgpu_fbdev_fini(adev);
@ -1627,8 +1635,7 @@ int amdgpu_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon)
 	struct amdgpu_device *adev;
 	struct drm_crtc *crtc;
 	struct drm_connector *connector;
-	int i, r;
-	bool force_completion = false;
+	int r;

 	if (dev == NULL || dev->dev_private == NULL) {
 		return -ENODEV;
@ -1667,21 +1674,7 @@ int amdgpu_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon)
 	/* evict vram memory */
 	amdgpu_bo_evict_vram(adev);

-	/* wait for gpu to finish processing current batch */
-	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
-		struct amdgpu_ring *ring = adev->rings[i];
-		if (!ring)
-			continue;
-
-		r = amdgpu_fence_wait_empty(ring);
-		if (r) {
-			/* delay GPU reset to resume */
-			force_completion = true;
-		}
-	}
-	if (force_completion) {
-		amdgpu_fence_driver_force_completion(adev);
-	}
+	amdgpu_fence_driver_suspend(adev);

 	r = amdgpu_suspend(adev);

@ -1739,6 +1732,8 @@ int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon)

 	r = amdgpu_resume(adev);

+	amdgpu_fence_driver_resume(adev);
+
 	r = amdgpu_ib_ring_tests(adev);
 	if (r)
 		DRM_ERROR("ib ring test failed (%d).\n", r);
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@ -35,6 +35,36 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>

+static void amdgpu_flip_wait_fence(struct amdgpu_device *adev,
+				   struct fence **f)
+{
+	struct amdgpu_fence *fence;
+	long r;
+
+	if (*f == NULL)
+		return;
+
+	fence = to_amdgpu_fence(*f);
+	if (fence) {
+		r = fence_wait(&fence->base, false);
+		if (r == -EDEADLK) {
+			up_read(&adev->exclusive_lock);
+			r = amdgpu_gpu_reset(adev);
+			down_read(&adev->exclusive_lock);
+		}
+	} else
+		r = fence_wait(*f, false);
+
+	if (r)
+		DRM_ERROR("failed to wait on page flip fence (%ld)!\n", r);
+
+	/* We continue with the page flip even if we failed to wait on
+	 * the fence, otherwise the DRM core and userspace will be
+	 * confused about which BO the CRTC is scanning out
+	 */
+	fence_put(*f);
+	*f = NULL;
+}

 static void amdgpu_flip_work_func(struct work_struct *__work)
 {
@ -44,34 +74,13 @@ static void amdgpu_flip_work_func(struct work_struct *__work)
 	struct amdgpu_crtc *amdgpuCrtc = adev->mode_info.crtcs[work->crtc_id];

 	struct drm_crtc *crtc = &amdgpuCrtc->base;
-	struct amdgpu_fence *fence;
 	unsigned long flags;
-	int r;
+	unsigned i;

 	down_read(&adev->exclusive_lock);
-	if (work->fence) {
-		fence = to_amdgpu_fence(work->fence);
-		if (fence) {
-			r = amdgpu_fence_wait(fence, false);
-			if (r == -EDEADLK) {
-				up_read(&adev->exclusive_lock);
-				r = amdgpu_gpu_reset(adev);
-				down_read(&adev->exclusive_lock);
-			}
-		} else
-			r = fence_wait(work->fence, false);
-
-		if (r)
-			DRM_ERROR("failed to wait on page flip fence (%d)!\n", r);
-
-		/* We continue with the page flip even if we failed to wait on
-		 * the fence, otherwise the DRM core and userspace will be
-		 * confused about which BO the CRTC is scanning out
-		 */
-
-		fence_put(work->fence);
-		work->fence = NULL;
-	}
+	amdgpu_flip_wait_fence(adev, &work->excl);
+	for (i = 0; i < work->shared_count; ++i)
+		amdgpu_flip_wait_fence(adev, &work->shared[i]);

 	/* We borrow the event spin lock for protecting flip_status */
 	spin_lock_irqsave(&crtc->dev->event_lock, flags);
@ -108,6 +117,7 @@ static void amdgpu_unpin_work_func(struct work_struct *__work)
 		DRM_ERROR("failed to reserve buffer after flip\n");

 	drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
+	kfree(work->shared);
 	kfree(work);
 }

@ -127,7 +137,7 @@ int amdgpu_crtc_page_flip(struct drm_crtc *crtc,
 	unsigned long flags;
 	u64 tiling_flags;
 	u64 base;
-	int r;
+	int i, r;

 	work = kzalloc(sizeof *work, GFP_KERNEL);
 	if (work == NULL)
@ -167,7 +177,19 @@ int amdgpu_crtc_page_flip(struct drm_crtc *crtc,
 		goto cleanup;
 	}

-	work->fence = fence_get(reservation_object_get_excl(new_rbo->tbo.resv));
+	r = reservation_object_get_fences_rcu(new_rbo->tbo.resv, &work->excl,
+					      &work->shared_count,
+					      &work->shared);
+	if (unlikely(r != 0)) {
+		amdgpu_bo_unreserve(new_rbo);
+		DRM_ERROR("failed to get fences for buffer\n");
+		goto cleanup;
+	}
+
+	fence_get(work->excl);
+	for (i = 0; i < work->shared_count; ++i)
+		fence_get(work->shared[i]);
+
 	amdgpu_bo_get_tiling_flags(new_rbo, &tiling_flags);
 	amdgpu_bo_unreserve(new_rbo);

@ -212,7 +234,10 @@ int amdgpu_crtc_page_flip(struct drm_crtc *crtc,

 cleanup:
 	drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
-	fence_put(work->fence);
+	fence_put(work->excl);
+	for (i = 0; i < work->shared_count; ++i)
+		fence_put(work->shared[i]);
+	kfree(work->shared);
 	kfree(work);

 	return r;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@ -63,7 +63,7 @@ int amdgpu_disp_priority = 0;
 int amdgpu_hw_i2c = 0;
 int amdgpu_pcie_gen2 = -1;
 int amdgpu_msi = -1;
-int amdgpu_lockup_timeout = 10000;
+int amdgpu_lockup_timeout = 0;
 int amdgpu_dpm = -1;
 int amdgpu_smc_load_fw = 1;
 int amdgpu_aspm = -1;
@ -75,6 +75,9 @@ int amdgpu_deep_color = 0;
 int amdgpu_vm_size = 8;
 int amdgpu_vm_block_size = -1;
 int amdgpu_exp_hw_support = 0;
+int amdgpu_enable_scheduler = 0;
+int amdgpu_sched_jobs = 16;
+int amdgpu_sched_hw_submission = 2;

 MODULE_PARM_DESC(vramlimit, "Restrict VRAM for testing, in megabytes");
 module_param_named(vramlimit, amdgpu_vram_limit, int, 0600);
@ -103,7 +106,7 @@ module_param_named(pcie_gen2, amdgpu_pcie_gen2, int, 0444);
 MODULE_PARM_DESC(msi, "MSI support (1 = enable, 0 = disable, -1 = auto)");
 module_param_named(msi, amdgpu_msi, int, 0444);

-MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (defaul 10000 = 10 seconds, 0 = disable)");
+MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default 0 = disable)");
 module_param_named(lockup_timeout, amdgpu_lockup_timeout, int, 0444);

 MODULE_PARM_DESC(dpm, "DPM support (1 = enable, 0 = disable, -1 = auto)");
@ -139,36 +142,45 @@ module_param_named(vm_block_size, amdgpu_vm_block_size, int, 0444);
 MODULE_PARM_DESC(exp_hw_support, "experimental hw support (1 = enable, 0 = disable (default))");
 module_param_named(exp_hw_support, amdgpu_exp_hw_support, int, 0444);

+MODULE_PARM_DESC(enable_scheduler, "enable SW GPU scheduler (1 = enable, 0 = disable ((default))");
+module_param_named(enable_scheduler, amdgpu_enable_scheduler, int, 0444);
+
+MODULE_PARM_DESC(sched_jobs, "the max number of jobs supported in the sw queue (default 16)");
+module_param_named(sched_jobs, amdgpu_sched_jobs, int, 0444);
+
+MODULE_PARM_DESC(sched_hw_submission, "the max number of HW submissions (default 2)");
+module_param_named(sched_hw_submission, amdgpu_sched_hw_submission, int, 0444);
+
 static struct pci_device_id pciidlist[] = {
 #ifdef CONFIG_DRM_AMDGPU_CIK
 	/* Kaveri */
-	{0x1002, 0x1304, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x1305, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1306, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x1307, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1309, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x130A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x130B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x130C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x130D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x130E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x130F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1310, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1311, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1312, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1313, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1315, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1316, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x1317, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x1318, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x131B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x131C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
-	{0x1002, 0x131D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMDGPU_IS_APU},
+	{0x1002, 0x1304, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x1305, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1306, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x1307, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1309, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x130A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x130B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x130C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x130D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x130E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x130F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1310, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1311, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1312, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1313, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1315, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1316, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x1317, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x1318, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x131B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x131C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
+	{0x1002, 0x131D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KAVERI|AMD_IS_APU},
 	/* Bonaire */
-	{0x1002, 0x6640, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMDGPU_IS_MOBILITY},
-	{0x1002, 0x6641, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMDGPU_IS_MOBILITY},
-	{0x1002, 0x6646, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMDGPU_IS_MOBILITY},
-	{0x1002, 0x6647, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMDGPU_IS_MOBILITY},
+	{0x1002, 0x6640, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMD_IS_MOBILITY},
+	{0x1002, 0x6641, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMD_IS_MOBILITY},
+	{0x1002, 0x6646, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMD_IS_MOBILITY},
+	{0x1002, 0x6647, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE|AMD_IS_MOBILITY},
 	{0x1002, 0x6649, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE},
 	{0x1002, 0x6650, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE},
 	{0x1002, 0x6651, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_BONAIRE},
@ -190,39 +202,39 @@ static struct pci_device_id pciidlist[] = {
 	{0x1002, 0x67BA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII},
 	{0x1002, 0x67BE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII},
 	/* Kabini */
-	{0x1002, 0x9830, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9831, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x9832, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9833, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x9834, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9835, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x9836, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9837, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x9838, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9839, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x983a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x983b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x983c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x983d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x983e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
-	{0x1002, 0x983f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMDGPU_IS_APU},
+	{0x1002, 0x9830, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9831, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x9832, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9833, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x9834, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9835, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x9836, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9837, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x9838, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9839, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x983a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x983b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x983c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x983d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x983e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
+	{0x1002, 0x983f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_KABINI|AMD_IS_APU},
 	/* mullins */
-	{0x1002, 0x9850, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9851, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9852, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9853, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9854, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9855, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9856, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9857, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9858, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x9859, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x985A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x985B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x985C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x985D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x985E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
-	{0x1002, 0x985F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMDGPU_IS_MOBILITY|AMDGPU_IS_APU},
+	{0x1002, 0x9850, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9851, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9852, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9853, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9854, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9855, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9856, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9857, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9858, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x9859, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x985A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x985B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x985C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x985D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x985E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
+	{0x1002, 0x985F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_MULLINS|AMD_IS_MOBILITY|AMD_IS_APU},
 #endif
 	/* topaz */
 	{0x1002, 0x6900, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TOPAZ},
@ -240,12 +252,14 @@ static struct pci_device_id pciidlist[] = {
 	{0x1002, 0x6930, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TONGA},
 	{0x1002, 0x6938, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TONGA},
 	{0x1002, 0x6939, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TONGA},
+	/* fiji */
+	{0x1002, 0x7300, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_FIJI},
 	/* carrizo */
-	{0x1002, 0x9870, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMDGPU_IS_APU},
-	{0x1002, 0x9874, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMDGPU_IS_APU},
-	{0x1002, 0x9875, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMDGPU_IS_APU},
-	{0x1002, 0x9876, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMDGPU_IS_APU},
-	{0x1002, 0x9877, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMDGPU_IS_APU},
+	{0x1002, 0x9870, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},
+	{0x1002, 0x9874, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},
+	{0x1002, 0x9875, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},
+	{0x1002, 0x9876, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},
+	{0x1002, 0x9877, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},

 	{0, 0, 0}
 };
@ -281,7 +295,7 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 	unsigned long flags = ent->driver_data;
 	int ret;

-	if ((flags & AMDGPU_EXP_HW_SUPPORT) && !amdgpu_exp_hw_support) {
+	if ((flags & AMD_EXP_HW_SUPPORT) && !amdgpu_exp_hw_support) {
 		DRM_INFO("This hardware requires experimental hardware support.\n"
 			 "See modparam exp_hw_support\n");
 		return -ENODEV;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h
@ -31,7 +31,7 @@
 #include <linux/firmware.h>
 #include <linux/platform_device.h>

-#include "amdgpu_family.h"
+#include "amd_shared.h"

 /* General customization:
 */
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
@ -53,9 +53,9 @@ static struct fb_ops amdgpufb_ops = {
 	.owner = THIS_MODULE,
 	.fb_check_var = drm_fb_helper_check_var,
 	.fb_set_par = drm_fb_helper_set_par,
-	.fb_fillrect = cfb_fillrect,
-	.fb_copyarea = cfb_copyarea,
-	.fb_imageblit = cfb_imageblit,
+	.fb_fillrect = drm_fb_helper_cfb_fillrect,
+	.fb_copyarea = drm_fb_helper_cfb_copyarea,
+	.fb_imageblit = drm_fb_helper_cfb_imageblit,
 	.fb_pan_display = drm_fb_helper_pan_display,
 	.fb_blank = drm_fb_helper_blank,
 	.fb_setcmap = drm_fb_helper_setcmap,
@ -179,7 +179,6 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	struct drm_mode_fb_cmd2 mode_cmd;
 	struct drm_gem_object *gobj = NULL;
 	struct amdgpu_bo *rbo = NULL;
-	struct device *device = &adev->pdev->dev;
 	int ret;
 	unsigned long tmp;

@ -201,9 +200,9 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	rbo = gem_to_amdgpu_bo(gobj);

 	/* okay we have an object now allocate the framebuffer */
-	info = framebuffer_alloc(0, device);
-	if (info == NULL) {
-		ret = -ENOMEM;
+	info = drm_fb_helper_alloc_fbi(helper);
+	if (IS_ERR(info)) {
+		ret = PTR_ERR(info);
 		goto out_unref;
 	}

@ -212,14 +211,13 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	ret = amdgpu_framebuffer_init(adev->ddev, &rfbdev->rfb, &mode_cmd, gobj);
 	if (ret) {
 		DRM_ERROR("failed to initialize framebuffer %d\n", ret);
-		goto out_unref;
+		goto out_destroy_fbi;
 	}

 	fb = &rfbdev->rfb.base;

 	/* setup helper */
 	rfbdev->helper.fb = fb;
-	rfbdev->helper.fbdev = info;

 	memset_io(rbo->kptr, 0x0, amdgpu_bo_size(rbo));

@ -239,11 +237,6 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	drm_fb_helper_fill_var(info, &rfbdev->helper, sizes->fb_width, sizes->fb_height);

 	/* setup aperture base/size for vesafb takeover */
-	info->apertures = alloc_apertures(1);
-	if (!info->apertures) {
-		ret = -ENOMEM;
-		goto out_unref;
-	}
 	info->apertures->ranges[0].base = adev->ddev->mode_config.fb_base;
 	info->apertures->ranges[0].size = adev->mc.aper_size;

@ -251,13 +244,7 @@ static int amdgpufb_create(struct drm_fb_helper *helper,

 	if (info->screen_base == NULL) {
 		ret = -ENOSPC;
-		goto out_unref;
-	}
-
-	ret = fb_alloc_cmap(&info->cmap, 256, 0);
-	if (ret) {
-		ret = -ENOMEM;
-		goto out_unref;
+		goto out_destroy_fbi;
 	}

 	DRM_INFO("fb mappable at 0x%lX\n",  info->fix.smem_start);
@ -269,6 +256,8 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	vga_switcheroo_client_fb_set(adev->ddev->pdev, info);
 	return 0;

+out_destroy_fbi:
+	drm_fb_helper_release_fbi(helper);
 out_unref:
 	if (rbo) {

@ -290,17 +279,10 @@ void amdgpu_fb_output_poll_changed(struct amdgpu_device *adev)

 static int amdgpu_fbdev_destroy(struct drm_device *dev, struct amdgpu_fbdev *rfbdev)
 {
-	struct fb_info *info;
 	struct amdgpu_framebuffer *rfb = &rfbdev->rfb;

-	if (rfbdev->helper.fbdev) {
-		info = rfbdev->helper.fbdev;
-
-		unregister_framebuffer(info);
-		if (info->cmap.len)
-			fb_dealloc_cmap(&info->cmap);
-		framebuffer_release(info);
-	}
+	drm_fb_helper_unregister_fbi(&rfbdev->helper);
+	drm_fb_helper_release_fbi(&rfbdev->helper);

 	if (rfb->obj) {
 		amdgpufb_destroy_pinned_object(rfb->obj);
@ -395,7 +377,8 @@ void amdgpu_fbdev_fini(struct amdgpu_device *adev)
 void amdgpu_fbdev_set_suspend(struct amdgpu_device *adev, int state)
 {
 	if (adev->mode_info.rfbdev)
-		fb_set_suspend(adev->mode_info.rfbdev->helper.fbdev, state);
+		drm_fb_helper_set_suspend(&adev->mode_info.rfbdev->helper,
+			state);
 }

 int amdgpu_fbdev_total_size(struct amdgpu_device *adev)
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@ -126,7 +126,8 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
 	(*fence)->ring = ring;
 	(*fence)->owner = owner;
 	fence_init(&(*fence)->base, &amdgpu_fence_ops,
-		&adev->fence_queue.lock, adev->fence_context + ring->idx,
+		&ring->fence_drv.fence_queue.lock,
+		adev->fence_context + ring->idx,
 		(*fence)->seq);
 	amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr,
 			       (*fence)->seq,
@ -135,38 +136,6 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
 	return 0;
 }

-/**
- * amdgpu_fence_recreate - recreate a fence from an user fence
- *
- * @ring: ring the fence is associated with
- * @owner: creator of the fence
- * @seq: user fence sequence number
- * @fence: resulting amdgpu fence object
- *
- * Recreates a fence command from the user fence sequence number (all asics).
- * Returns 0 on success, -ENOMEM on failure.
- */
-int amdgpu_fence_recreate(struct amdgpu_ring *ring, void *owner,
-			  uint64_t seq, struct amdgpu_fence **fence)
-{
-	struct amdgpu_device *adev = ring->adev;
-
-	if (seq > ring->fence_drv.sync_seq[ring->idx])
-		return -EINVAL;
-
-	*fence = kmalloc(sizeof(struct amdgpu_fence), GFP_KERNEL);
-	if ((*fence) == NULL)
-		return -ENOMEM;
-
-	(*fence)->seq = seq;
-	(*fence)->ring = ring;
-	(*fence)->owner = owner;
-	fence_init(&(*fence)->base, &amdgpu_fence_ops,
-		&adev->fence_queue.lock, adev->fence_context + ring->idx,
-		(*fence)->seq);
-	return 0;
-}
-
 /**
 * amdgpu_fence_check_signaled - callback from fence_queue
 *
@ -196,9 +165,7 @@ static int amdgpu_fence_check_signaled(wait_queue_t *wait, unsigned mode, int fl
 		else
 			FENCE_TRACE(&fence->base, "was already signaled\n");

-		amdgpu_irq_put(adev, fence->ring->fence_drv.irq_src,
-				fence->ring->fence_drv.irq_type);
-		__remove_wait_queue(&adev->fence_queue, &fence->fence_wake);
+		__remove_wait_queue(&fence->ring->fence_drv.fence_queue, &fence->fence_wake);
 		fence_put(&fence->base);
 	} else
 		FENCE_TRACE(&fence->base, "pending\n");
@ -299,14 +266,9 @@ static void amdgpu_fence_check_lockup(struct work_struct *work)
 		return;
 	}

-	if (fence_drv->delayed_irq && ring->adev->ddev->irq_enabled) {
-		fence_drv->delayed_irq = false;
-		amdgpu_irq_update(ring->adev, fence_drv->irq_src,
-				fence_drv->irq_type);
+	if (amdgpu_fence_activity(ring)) {
+		wake_up_all(&ring->fence_drv.fence_queue);
 	}
-
-	if (amdgpu_fence_activity(ring))
-		wake_up_all(&ring->adev->fence_queue);
 	else if (amdgpu_ring_is_lockup(ring)) {
 		/* good news we believe it's a lockup */
 		dev_warn(ring->adev->dev, "GPU lockup (current fence id "
@ -316,7 +278,7 @@ static void amdgpu_fence_check_lockup(struct work_struct *work)

 		/* remember that we need an reset */
 		ring->adev->needs_reset = true;
-		wake_up_all(&ring->adev->fence_queue);
+		wake_up_all(&ring->fence_drv.fence_queue);
 	}
 	up_read(&ring->adev->exclusive_lock);
 }
@ -332,62 +294,8 @@ static void amdgpu_fence_check_lockup(struct work_struct *work)
 */
 void amdgpu_fence_process(struct amdgpu_ring *ring)
 {
-	uint64_t seq, last_seq, last_emitted;
-	unsigned count_loop = 0;
-	bool wake = false;
-
-	/* Note there is a scenario here for an infinite loop but it's
-	 * very unlikely to happen. For it to happen, the current polling
-	 * process need to be interrupted by another process and another
-	 * process needs to update the last_seq btw the atomic read and
-	 * xchg of the current process.
-	 *
-	 * More over for this to go in infinite loop there need to be
-	 * continuously new fence signaled ie amdgpu_fence_read needs
-	 * to return a different value each time for both the currently
-	 * polling process and the other process that xchg the last_seq
-	 * btw atomic read and xchg of the current process. And the
-	 * value the other process set as last seq must be higher than
-	 * the seq value we just read. Which means that current process
-	 * need to be interrupted after amdgpu_fence_read and before
-	 * atomic xchg.
-	 *
-	 * To be even more safe we count the number of time we loop and
-	 * we bail after 10 loop just accepting the fact that we might
-	 * have temporarly set the last_seq not to the true real last
-	 * seq but to an older one.
-	 */
-	last_seq = atomic64_read(&ring->fence_drv.last_seq);
-	do {
-		last_emitted = ring->fence_drv.sync_seq[ring->idx];
-		seq = amdgpu_fence_read(ring);
-		seq |= last_seq & 0xffffffff00000000LL;
-		if (seq < last_seq) {
-			seq &= 0xffffffff;
-			seq |= last_emitted & 0xffffffff00000000LL;
-		}
-
-		if (seq <= last_seq || seq > last_emitted) {
-			break;
-		}
-		/* If we loop over we don't want to return without
-		 * checking if a fence is signaled as it means that the
-		 * seq we just read is different from the previous on.
-		 */
-		wake = true;
-		last_seq = seq;
-		if ((count_loop++) > 10) {
-			/* We looped over too many time leave with the
-			 * fact that we might have set an older fence
-			 * seq then the current real last seq as signaled
-			 * by the hw.
-			 */
-			break;
-		}
-	} while (atomic64_xchg(&ring->fence_drv.last_seq, seq) > seq);
-
-	if (wake)
-		wake_up_all(&ring->adev->fence_queue);
+	if (amdgpu_fence_activity(ring))
+		wake_up_all(&ring->fence_drv.fence_queue);
 }

 /**
@ -447,284 +355,49 @@ static bool amdgpu_fence_enable_signaling(struct fence *f)
 {
 	struct amdgpu_fence *fence = to_amdgpu_fence(f);
 	struct amdgpu_ring *ring = fence->ring;
-	struct amdgpu_device *adev = ring->adev;

 	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
 		return false;

-	if (down_read_trylock(&adev->exclusive_lock)) {
-		amdgpu_irq_get(adev, ring->fence_drv.irq_src,
-			ring->fence_drv.irq_type);
-		if (amdgpu_fence_activity(ring))
-			wake_up_all_locked(&adev->fence_queue);
-
-		/* did fence get signaled after we enabled the sw irq? */
-		if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq) {
-			amdgpu_irq_put(adev, ring->fence_drv.irq_src,
-				ring->fence_drv.irq_type);
-			up_read(&adev->exclusive_lock);
-			return false;
-		}
-
-		up_read(&adev->exclusive_lock);
-	} else {
-		/* we're probably in a lockup, lets not fiddle too much */
-		if (amdgpu_irq_get_delayed(adev, ring->fence_drv.irq_src,
-			ring->fence_drv.irq_type))
-			ring->fence_drv.delayed_irq = true;
-		amdgpu_fence_schedule_check(ring);
-	}
-
 	fence->fence_wake.flags = 0;
 	fence->fence_wake.private = NULL;
 	fence->fence_wake.func = amdgpu_fence_check_signaled;
-	__add_wait_queue(&adev->fence_queue, &fence->fence_wake);
+	__add_wait_queue(&ring->fence_drv.fence_queue, &fence->fence_wake);
 	fence_get(f);
 	FENCE_TRACE(&fence->base, "armed on ring %i!\n", ring->idx);
 	return true;
 }

-/**
- * amdgpu_fence_signaled - check if a fence has signaled
+/*
+ * amdgpu_ring_wait_seq_timeout - wait for seq of the specific ring to signal
+ * @ring: ring to wait on for the seq number
+ * @seq: seq number wait for
 *
- * @fence: amdgpu fence object
- *
- * Check if the requested fence has signaled (all asics).
- * Returns true if the fence has signaled or false if it has not.
+ * return value:
+ * 0: seq signaled, and gpu not hang
+ * -EDEADL: GPU hang detected
+ * -EINVAL: some paramter is not valid
 */
-bool amdgpu_fence_signaled(struct amdgpu_fence *fence)
+static int amdgpu_fence_ring_wait_seq(struct amdgpu_ring *ring, uint64_t seq)
 {
-	if (!fence)
-		return true;
+	struct amdgpu_device *adev = ring->adev;
+	bool signaled = false;

-	if (amdgpu_fence_seq_signaled(fence->ring, fence->seq)) {
-		if (!fence_signal(&fence->base))
-			FENCE_TRACE(&fence->base, "signaled from amdgpu_fence_signaled\n");
-		return true;
-	}
+	BUG_ON(!ring);
+	if (seq > ring->fence_drv.sync_seq[ring->idx])
+		return -EINVAL;

-	return false;
-}
+	if (atomic64_read(&ring->fence_drv.last_seq) >= seq)
+		return 0;

-/**
- * amdgpu_fence_any_seq_signaled - check if any sequence number is signaled
- *
- * @adev: amdgpu device pointer
- * @seq: sequence numbers
- *
- * Check if the last signaled fence sequnce number is >= the requested
- * sequence number (all asics).
- * Returns true if any has signaled (current value is >= requested value)
- * or false if it has not. Helper function for amdgpu_fence_wait_seq.
- */
-static bool amdgpu_fence_any_seq_signaled(struct amdgpu_device *adev, u64 *seq)
-{
-	unsigned i;
+	wait_event(ring->fence_drv.fence_queue, (
+		   (signaled = amdgpu_fence_seq_signaled(ring, seq))
+		   || adev->needs_reset));

-	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		if (!adev->rings[i] || !seq[i])
-			continue;
-
-		if (amdgpu_fence_seq_signaled(adev->rings[i], seq[i]))
-			return true;
-	}
-
-	return false;
-}
-
-/**
- * amdgpu_fence_wait_seq_timeout - wait for a specific sequence numbers
- *
- * @adev: amdgpu device pointer
- * @target_seq: sequence number(s) we want to wait for
- * @intr: use interruptable sleep
- * @timeout: maximum time to wait, or MAX_SCHEDULE_TIMEOUT for infinite wait
- *
- * Wait for the requested sequence number(s) to be written by any ring
- * (all asics).  Sequnce number array is indexed by ring id.
- * @intr selects whether to use interruptable (true) or non-interruptable
- * (false) sleep when waiting for the sequence number.  Helper function
- * for amdgpu_fence_wait_*().
- * Returns remaining time if the sequence number has passed, 0 when
- * the wait timeout, or an error for all other cases.
- * -EDEADLK is returned when a GPU lockup has been detected.
- */
-static long amdgpu_fence_wait_seq_timeout(struct amdgpu_device *adev,
-					  u64 *target_seq, bool intr,
-					  long timeout)
-{
-	uint64_t last_seq[AMDGPU_MAX_RINGS];
-	bool signaled;
-	int i;
-	long r;
-
-	if (timeout == 0) {
-		return amdgpu_fence_any_seq_signaled(adev, target_seq);
-	}
-
-	while (!amdgpu_fence_any_seq_signaled(adev, target_seq)) {
-
-		/* Save current sequence values, used to check for GPU lockups */
-		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-			struct amdgpu_ring *ring = adev->rings[i];
-
-			if (!ring || !target_seq[i])
-				continue;
-
-			last_seq[i] = atomic64_read(&ring->fence_drv.last_seq);
-			trace_amdgpu_fence_wait_begin(adev->ddev, i, target_seq[i]);
-			amdgpu_irq_get(adev, ring->fence_drv.irq_src,
-				       ring->fence_drv.irq_type);
-		}
-
-		if (intr) {
-			r = wait_event_interruptible_timeout(adev->fence_queue, (
-				(signaled = amdgpu_fence_any_seq_signaled(adev, target_seq))
-				 || adev->needs_reset), AMDGPU_FENCE_JIFFIES_TIMEOUT);
-		} else {
-			r = wait_event_timeout(adev->fence_queue, (
-				(signaled = amdgpu_fence_any_seq_signaled(adev, target_seq))
-				 || adev->needs_reset), AMDGPU_FENCE_JIFFIES_TIMEOUT);
-		}
-
-		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-			struct amdgpu_ring *ring = adev->rings[i];
-
-			if (!ring || !target_seq[i])
-				continue;
-
-			amdgpu_irq_put(adev, ring->fence_drv.irq_src,
-				       ring->fence_drv.irq_type);
-			trace_amdgpu_fence_wait_end(adev->ddev, i, target_seq[i]);
-		}
-
-		if (unlikely(r < 0))
-			return r;
-
-		if (unlikely(!signaled)) {
-
-			if (adev->needs_reset)
-				return -EDEADLK;
-
-			/* we were interrupted for some reason and fence
-			 * isn't signaled yet, resume waiting */
-			if (r)
-				continue;
-
-			for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-				struct amdgpu_ring *ring = adev->rings[i];
-
-				if (!ring || !target_seq[i])
-					continue;
-
-				if (last_seq[i] != atomic64_read(&ring->fence_drv.last_seq))
-					break;
-			}
-
-			if (i != AMDGPU_MAX_RINGS)
-				continue;
-
-			for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-				if (!adev->rings[i] || !target_seq[i])
-					continue;
-
-				if (amdgpu_ring_is_lockup(adev->rings[i]))
-					break;
-			}
-
-			if (i < AMDGPU_MAX_RINGS) {
-				/* good news we believe it's a lockup */
-				dev_warn(adev->dev, "GPU lockup (waiting for "
-					 "0x%016llx last fence id 0x%016llx on"
-					 " ring %d)\n",
-					 target_seq[i], last_seq[i], i);
-
-				/* remember that we need an reset */
-				adev->needs_reset = true;
-				wake_up_all(&adev->fence_queue);
-				return -EDEADLK;
-			}
-
-			if (timeout < MAX_SCHEDULE_TIMEOUT) {
-				timeout -= AMDGPU_FENCE_JIFFIES_TIMEOUT;
-				if (timeout <= 0) {
-					return 0;
-				}
-			}
-		}
-	}
-	return timeout;
-}
-
-/**
- * amdgpu_fence_wait - wait for a fence to signal
- *
- * @fence: amdgpu fence object
- * @intr: use interruptable sleep
- *
- * Wait for the requested fence to signal (all asics).
- * @intr selects whether to use interruptable (true) or non-interruptable
- * (false) sleep when waiting for the fence.
- * Returns 0 if the fence has passed, error for all other cases.
- */
-int amdgpu_fence_wait(struct amdgpu_fence *fence, bool intr)
-{
-	uint64_t seq[AMDGPU_MAX_RINGS] = {};
-	long r;
-
-	seq[fence->ring->idx] = fence->seq;
-	r = amdgpu_fence_wait_seq_timeout(fence->ring->adev, seq, intr, MAX_SCHEDULE_TIMEOUT);
-	if (r < 0) {
-		return r;
-	}
-
-	r = fence_signal(&fence->base);
-	if (!r)
-		FENCE_TRACE(&fence->base, "signaled from fence_wait\n");
-	return 0;
-}
-
-/**
- * amdgpu_fence_wait_any - wait for a fence to signal on any ring
- *
- * @adev: amdgpu device pointer
- * @fences: amdgpu fence object(s)
- * @intr: use interruptable sleep
- *
- * Wait for any requested fence to signal (all asics).  Fence
- * array is indexed by ring id.  @intr selects whether to use
- * interruptable (true) or non-interruptable (false) sleep when
- * waiting for the fences. Used by the suballocator.
- * Returns 0 if any fence has passed, error for all other cases.
- */
-int amdgpu_fence_wait_any(struct amdgpu_device *adev,
-			  struct amdgpu_fence **fences,
-			  bool intr)
-{
-	uint64_t seq[AMDGPU_MAX_RINGS];
-	unsigned i, num_rings = 0;
-	long r;
-
-	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		seq[i] = 0;
-
-		if (!fences[i]) {
-			continue;
-		}
-
-		seq[i] = fences[i]->seq;
-		++num_rings;
-	}
-
-	/* nothing to wait for ? */
-	if (num_rings == 0)
-		return -ENOENT;
-
-	r = amdgpu_fence_wait_seq_timeout(adev, seq, intr, MAX_SCHEDULE_TIMEOUT);
-	if (r < 0) {
-		return r;
-	}
-	return 0;
+	if (signaled)
+		return 0;
+	else
+		return -EDEADLK;
 }

 /**
@ -739,19 +412,12 @@ int amdgpu_fence_wait_any(struct amdgpu_device *adev,
 */
 int amdgpu_fence_wait_next(struct amdgpu_ring *ring)
 {
-	uint64_t seq[AMDGPU_MAX_RINGS] = {};
-	long r;
+	uint64_t seq = atomic64_read(&ring->fence_drv.last_seq) + 1ULL;

-	seq[ring->idx] = atomic64_read(&ring->fence_drv.last_seq) + 1ULL;
-	if (seq[ring->idx] >= ring->fence_drv.sync_seq[ring->idx]) {
-		/* nothing to wait for, last_seq is
-		   already the last emited fence */
+	if (seq >= ring->fence_drv.sync_seq[ring->idx])
 		return -ENOENT;
-	}
-	r = amdgpu_fence_wait_seq_timeout(ring->adev, seq, false, MAX_SCHEDULE_TIMEOUT);
-	if (r < 0)
-		return r;
-	return 0;
+
+	return amdgpu_fence_ring_wait_seq(ring, seq);
 }

 /**
@ -766,23 +432,12 @@ int amdgpu_fence_wait_next(struct amdgpu_ring *ring)
 */
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring)
 {
-	struct amdgpu_device *adev = ring->adev;
-	uint64_t seq[AMDGPU_MAX_RINGS] = {};
-	long r;
+	uint64_t seq = ring->fence_drv.sync_seq[ring->idx];

-	seq[ring->idx] = ring->fence_drv.sync_seq[ring->idx];
-	if (!seq[ring->idx])
+	if (!seq)
 		return 0;

-	r = amdgpu_fence_wait_seq_timeout(adev, seq, false, MAX_SCHEDULE_TIMEOUT);
-	if (r < 0) {
-		if (r == -EDEADLK)
-			return -EDEADLK;
-
-		dev_err(adev->dev, "error waiting for ring[%d] to become idle (%ld)\n",
-			ring->idx, r);
-	}
-	return 0;
+	return amdgpu_fence_ring_wait_seq(ring, seq);
 }

 /**
@ -933,9 +588,12 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
 		ring->fence_drv.gpu_addr = adev->uvd.gpu_addr + index;
 	}
 	amdgpu_fence_write(ring, atomic64_read(&ring->fence_drv.last_seq));
-	ring->fence_drv.initialized = true;
+	amdgpu_irq_get(adev, irq_src, irq_type);
+
 	ring->fence_drv.irq_src = irq_src;
 	ring->fence_drv.irq_type = irq_type;
+	ring->fence_drv.initialized = true;
+
 	dev_info(adev->dev, "fence driver on ring %d use gpu addr 0x%016llx, "
 		 "cpu addr 0x%p\n", ring->idx,
 		 ring->fence_drv.gpu_addr, ring->fence_drv.cpu_addr);
@ -966,6 +624,16 @@ void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring)
 	INIT_DELAYED_WORK(&ring->fence_drv.lockup_work,
 			amdgpu_fence_check_lockup);
 	ring->fence_drv.ring = ring;
+
+	if (amdgpu_enable_scheduler) {
+		ring->scheduler = amd_sched_create((void *)ring->adev,
+						   &amdgpu_sched_ops,
+						   ring->idx, 5, 0,
+						   amdgpu_sched_hw_submission);
+		if (!ring->scheduler)
+			DRM_ERROR("Failed to create scheduler on ring %d.\n",
+				  ring->idx);
+	}
 }

 /**
@ -982,7 +650,6 @@ void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring)
 */
 int amdgpu_fence_driver_init(struct amdgpu_device *adev)
 {
-	init_waitqueue_head(&adev->fence_queue);
 	if (amdgpu_debugfs_fence_init(adev))
 		dev_err(adev->dev, "fence debugfs file creation failed\n");

@ -1011,12 +678,77 @@ void amdgpu_fence_driver_fini(struct amdgpu_device *adev)
 			/* no need to trigger GPU reset as we are unloading */
 			amdgpu_fence_driver_force_completion(adev);
 		}
-		wake_up_all(&adev->fence_queue);
+		wake_up_all(&ring->fence_drv.fence_queue);
+		amdgpu_irq_put(adev, ring->fence_drv.irq_src,
+			       ring->fence_drv.irq_type);
+		if (ring->scheduler)
+			amd_sched_destroy(ring->scheduler);
 		ring->fence_drv.initialized = false;
 	}
 	mutex_unlock(&adev->ring_lock);
 }

+/**
+ * amdgpu_fence_driver_suspend - suspend the fence driver
+ * for all possible rings.
+ *
+ * @adev: amdgpu device pointer
+ *
+ * Suspend the fence driver for all possible rings (all asics).
+ */
+void amdgpu_fence_driver_suspend(struct amdgpu_device *adev)
+{
+	int i, r;
+
+	mutex_lock(&adev->ring_lock);
+	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
+		struct amdgpu_ring *ring = adev->rings[i];
+		if (!ring || !ring->fence_drv.initialized)
+			continue;
+
+		/* wait for gpu to finish processing current batch */
+		r = amdgpu_fence_wait_empty(ring);
+		if (r) {
+			/* delay GPU reset to resume */
+			amdgpu_fence_driver_force_completion(adev);
+		}
+
+		/* disable the interrupt */
+		amdgpu_irq_put(adev, ring->fence_drv.irq_src,
+			       ring->fence_drv.irq_type);
+	}
+	mutex_unlock(&adev->ring_lock);
+}
+
+/**
+ * amdgpu_fence_driver_resume - resume the fence driver
+ * for all possible rings.
+ *
+ * @adev: amdgpu device pointer
+ *
+ * Resume the fence driver for all possible rings (all asics).
+ * Not all asics have all rings, so each asic will only
+ * start the fence driver on the rings it has using
+ * amdgpu_fence_driver_start_ring().
+ * Returns 0 for success.
+ */
+void amdgpu_fence_driver_resume(struct amdgpu_device *adev)
+{
+	int i;
+
+	mutex_lock(&adev->ring_lock);
+	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
+		struct amdgpu_ring *ring = adev->rings[i];
+		if (!ring || !ring->fence_drv.initialized)
+			continue;
+
+		/* enable the interrupt */
+		amdgpu_irq_get(adev, ring->fence_drv.irq_src,
+			       ring->fence_drv.irq_type);
+	}
+	mutex_unlock(&adev->ring_lock);
+}
+
 /**
 * amdgpu_fence_driver_force_completion - force all fence waiter to complete
 *
@ -1104,6 +836,22 @@ static inline bool amdgpu_test_signaled(struct amdgpu_fence *fence)
 	return test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->base.flags);
 }

+static inline bool amdgpu_test_signaled_any(struct amdgpu_fence **fences)
+{
+	int idx;
+	struct amdgpu_fence *fence;
+
+	idx = 0;
+	for (idx = 0; idx < AMDGPU_MAX_RINGS; ++idx) {
+		fence = fences[idx];
+		if (fence) {
+			if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->base.flags))
+				return true;
+		}
+	}
+	return false;
+}
+
 struct amdgpu_wait_cb {
 	struct fence_cb base;
 	struct task_struct *task;
@ -1119,14 +867,35 @@ static void amdgpu_fence_wait_cb(struct fence *fence, struct fence_cb *cb)
 static signed long amdgpu_fence_default_wait(struct fence *f, bool intr,
 					     signed long t)
 {
+	struct amdgpu_fence *array[AMDGPU_MAX_RINGS];
 	struct amdgpu_fence *fence = to_amdgpu_fence(f);
 	struct amdgpu_device *adev = fence->ring->adev;
-	struct amdgpu_wait_cb cb;

-	cb.task = current;
+	memset(&array[0], 0, sizeof(array));
+	array[0] = fence;

-	if (fence_add_callback(f, &cb.base, amdgpu_fence_wait_cb))
-		return t;
+	return amdgpu_fence_wait_any(adev, array, intr, t);
+}
+
+/* wait until any fence in array signaled */
+signed long amdgpu_fence_wait_any(struct amdgpu_device *adev,
+				struct amdgpu_fence **array, bool intr, signed long t)
+{
+	long idx = 0;
+	struct amdgpu_wait_cb cb[AMDGPU_MAX_RINGS];
+	struct amdgpu_fence *fence;
+
+	BUG_ON(!array);
+
+	for (idx = 0; idx < AMDGPU_MAX_RINGS; ++idx) {
+		fence = array[idx];
+		if (fence) {
+			cb[idx].task = current;
+			if (fence_add_callback(&fence->base,
+					&cb[idx].base, amdgpu_fence_wait_cb))
+				return t; /* return if fence is already signaled */
+		}
+	}

 	while (t > 0) {
 		if (intr)
@ -1135,10 +904,10 @@ static signed long amdgpu_fence_default_wait(struct fence *f, bool intr,
 			set_current_state(TASK_UNINTERRUPTIBLE);

 		/*
-		 * amdgpu_test_signaled must be called after
+		 * amdgpu_test_signaled_any must be called after
 		 * set_current_state to prevent a race with wake_up_process
 		 */
-		if (amdgpu_test_signaled(fence))
+		if (amdgpu_test_signaled_any(array))
 			break;

 		if (adev->needs_reset) {
@ -1153,7 +922,13 @@ static signed long amdgpu_fence_default_wait(struct fence *f, bool intr,
 	}

 	__set_current_state(TASK_RUNNING);
-	fence_remove_callback(f, &cb.base);
+
+	idx = 0;
+	for (idx = 0; idx < AMDGPU_MAX_RINGS; ++idx) {
+		fence = array[idx];
+		if (fence)
+			fence_remove_callback(&fence->base, &cb[idx].base);
+	}

 	return t;
 }
--- a/Show More
+++ b/Show More