Merge drm/drm-next into drm-misc-next

Backmerging drm-next to sync with msm tree. Resolves a conflict between aperture-helper changes and msm's use of those interfaces. Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
2023-04-16 18:50:39 +02:00 · 2023-04-16 18:50:39 +02:00 · 7f6f26d7ad
commit 7f6f26d7ad
parent 116b1c5a36 afa351a15d
446 changed files with 134074 additions and 9978 deletions
--- a/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon
+++ b/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon
@ -14,7 +14,9 @@ Description:	RW. Card reactive sustained  (PL1/Tau) power limit in microwatts.

 		The power controller will throttle the operating frequency
 		if the power averaged over a window (typically seconds)
-		exceeds this limit.
+		exceeds this limit. A read value of 0 means that the PL1
+		power limit is disabled, writing 0 disables the
+		limit. Writing values > 0 will enable the power limit.

 		Only supported for particular Intel i915 graphics platforms.

--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,ccorr.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,ccorr.yaml
@ -27,13 +27,10 @@ properties:
          - const: mediatek,mt8192-disp-ccorr
      - items:
          - enum:
+              - mediatek,mt8186-disp-ccorr
              - mediatek,mt8188-disp-ccorr
              - mediatek,mt8195-disp-ccorr
          - const: mediatek,mt8192-disp-ccorr
-      - items:
-          - enum:
-              - mediatek,mt8186-disp-ccorr
-          - const: mediatek,mt8192-disp-ccorr

  reg:
    maxItems: 1
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,ethdr.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,ethdr.yaml
@ -0,0 +1,182 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/mediatek/mediatek,ethdr.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: MediaTek Ethdr Device
+
+maintainers:
+  - Chun-Kuang Hu <chunkuang.hu@kernel.org>
+  - Philipp Zabel <p.zabel@pengutronix.de>
+
+description:
+  ETHDR (ET High Dynamic Range) is a MediaTek internal HDR engine and is
+  designed for HDR video and graphics conversion in the external display path.
+  It handles multiple HDR input types and performs tone mapping, color
+  space/color format conversion, and then combine different layers,
+  output the required HDR or SDR signal to the subsequent display path.
+  This engine is composed of two video frontends, two graphic frontends,
+  one video backend and a mixer. ETHDR has two DMA function blocks, DS and ADL.
+  These two function blocks read the pre-programmed registers from DRAM and
+  set them to HW in the v-blanking period.
+
+properties:
+  compatible:
+    const: mediatek,mt8195-disp-ethdr
+
+  reg:
+    maxItems: 7
+
+  reg-names:
+    items:
+      - const: mixer
+      - const: vdo_fe0
+      - const: vdo_fe1
+      - const: gfx_fe0
+      - const: gfx_fe1
+      - const: vdo_be
+      - const: adl_ds
+
+  interrupts:
+    maxItems: 1
+
+  iommus:
+    minItems: 1
+    maxItems: 2
+
+  clocks:
+    items:
+      - description: mixer clock
+      - description: video frontend 0 clock
+      - description: video frontend 1 clock
+      - description: graphic frontend 0 clock
+      - description: graphic frontend 1 clock
+      - description: video backend clock
+      - description: autodownload and menuload clock
+      - description: video frontend 0 async clock
+      - description: video frontend 1 async clock
+      - description: graphic frontend 0 async clock
+      - description: graphic frontend 1 async clock
+      - description: video backend async clock
+      - description: ethdr top clock
+
+  clock-names:
+    items:
+      - const: mixer
+      - const: vdo_fe0
+      - const: vdo_fe1
+      - const: gfx_fe0
+      - const: gfx_fe1
+      - const: vdo_be
+      - const: adl_ds
+      - const: vdo_fe0_async
+      - const: vdo_fe1_async
+      - const: gfx_fe0_async
+      - const: gfx_fe1_async
+      - const: vdo_be_async
+      - const: ethdr_top
+
+  power-domains:
+    maxItems: 1
+
+  resets:
+    items:
+      - description: video frontend 0 async reset
+      - description: video frontend 1 async reset
+      - description: graphic frontend 0 async reset
+      - description: graphic frontend 1 async reset
+      - description: video backend async reset
+
+  reset-names:
+    items:
+      - const: vdo_fe0_async
+      - const: vdo_fe1_async
+      - const: gfx_fe0_async
+      - const: gfx_fe1_async
+      - const: vdo_be_async
+
+  mediatek,gce-client-reg:
+    $ref: /schemas/types.yaml#/definitions/phandle-array
+    minItems: 1
+    maxItems: 7
+    description: The register of display function block to be set by gce.
+      There are 4 arguments in this property, gce node, subsys id, offset and
+      register size. The subsys id is defined in the gce header of each chips
+      include/dt-bindings/gce/<chip>-gce.h, mapping to the register of display
+      function block.
+
+required:
+  - compatible
+  - reg
+  - clocks
+  - clock-names
+  - interrupts
+  - power-domains
+  - resets
+  - mediatek,gce-client-reg
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/clock/mt8195-clk.h>
+    #include <dt-bindings/gce/mt8195-gce.h>
+    #include <dt-bindings/memory/mt8195-memory-port.h>
+    #include <dt-bindings/power/mt8195-power.h>
+    #include <dt-bindings/reset/mt8195-resets.h>
+
+    soc {
+        #address-cells = <2>;
+        #size-cells = <2>;
+
+        hdr-engine@1c114000 {
+                compatible = "mediatek,mt8195-disp-ethdr";
+                reg = <0 0x1c114000 0 0x1000>,
+                      <0 0x1c115000 0 0x1000>,
+                      <0 0x1c117000 0 0x1000>,
+                      <0 0x1c119000 0 0x1000>,
+                      <0 0x1c11a000 0 0x1000>,
+                      <0 0x1c11b000 0 0x1000>,
+                      <0 0x1c11c000 0 0x1000>;
+                reg-names = "mixer", "vdo_fe0", "vdo_fe1", "gfx_fe0", "gfx_fe1",
+                            "vdo_be", "adl_ds";
+                mediatek,gce-client-reg = <&gce0 SUBSYS_1c11XXXX 0x4000 0x1000>,
+                                          <&gce0 SUBSYS_1c11XXXX 0x5000 0x1000>,
+                                          <&gce0 SUBSYS_1c11XXXX 0x7000 0x1000>,
+                                          <&gce0 SUBSYS_1c11XXXX 0x9000 0x1000>,
+                                          <&gce0 SUBSYS_1c11XXXX 0xa000 0x1000>,
+                                          <&gce0 SUBSYS_1c11XXXX 0xb000 0x1000>,
+                                          <&gce0 SUBSYS_1c11XXXX 0xc000 0x1000>;
+                clocks = <&vdosys1 CLK_VDO1_DISP_MIXER>,
+                         <&vdosys1 CLK_VDO1_HDR_VDO_FE0>,
+                         <&vdosys1 CLK_VDO1_HDR_VDO_FE1>,
+                         <&vdosys1 CLK_VDO1_HDR_GFX_FE0>,
+                         <&vdosys1 CLK_VDO1_HDR_GFX_FE1>,
+                         <&vdosys1 CLK_VDO1_HDR_VDO_BE>,
+                         <&vdosys1 CLK_VDO1_26M_SLOW>,
+                         <&vdosys1 CLK_VDO1_HDR_VDO_FE0_DL_ASYNC>,
+                         <&vdosys1 CLK_VDO1_HDR_VDO_FE1_DL_ASYNC>,
+                         <&vdosys1 CLK_VDO1_HDR_GFX_FE0_DL_ASYNC>,
+                         <&vdosys1 CLK_VDO1_HDR_GFX_FE1_DL_ASYNC>,
+                         <&vdosys1 CLK_VDO1_HDR_VDO_BE_DL_ASYNC>,
+                         <&topckgen CLK_TOP_ETHDR>;
+                clock-names = "mixer", "vdo_fe0", "vdo_fe1", "gfx_fe0", "gfx_fe1",
+                              "vdo_be", "adl_ds", "vdo_fe0_async", "vdo_fe1_async",
+                              "gfx_fe0_async", "gfx_fe1_async","vdo_be_async",
+                              "ethdr_top";
+                power-domains = <&spm MT8195_POWER_DOMAIN_VDOSYS1>;
+                iommus = <&iommu_vpp M4U_PORT_L3_HDR_DS>,
+                         <&iommu_vpp M4U_PORT_L3_HDR_ADL>;
+                interrupts = <GIC_SPI 517 IRQ_TYPE_LEVEL_HIGH 0>; /* disp mixer */
+                resets = <&vdosys1 MT8195_VDOSYS1_SW1_RST_B_HDR_VDO_FE0_DL_ASYNC>,
+                         <&vdosys1 MT8195_VDOSYS1_SW1_RST_B_HDR_VDO_FE1_DL_ASYNC>,
+                         <&vdosys1 MT8195_VDOSYS1_SW1_RST_B_HDR_GFX_FE0_DL_ASYNC>,
+                         <&vdosys1 MT8195_VDOSYS1_SW1_RST_B_HDR_GFX_FE1_DL_ASYNC>,
+                         <&vdosys1 MT8195_VDOSYS1_SW1_RST_B_HDR_VDO_BE_DL_ASYNC>;
+                reset-names = "vdo_fe0_async", "vdo_fe1_async", "gfx_fe0_async",
+                              "gfx_fe1_async", "vdo_be_async";
+        };
+    };
+...
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@ -15,16 +15,21 @@ description: |

 properties:
  compatible:
-    enum:
-      - qcom,sc7180-dp
-      - qcom,sc7280-dp
-      - qcom,sc7280-edp
-      - qcom,sc8180x-dp
-      - qcom,sc8180x-edp
-      - qcom,sc8280xp-dp
-      - qcom,sc8280xp-edp
-      - qcom,sdm845-dp
-      - qcom,sm8350-dp
+    oneOf:
+      - enum:
+          - qcom,sc7180-dp
+          - qcom,sc7280-dp
+          - qcom,sc7280-edp
+          - qcom,sc8180x-dp
+          - qcom,sc8180x-edp
+          - qcom,sc8280xp-dp
+          - qcom,sc8280xp-edp
+          - qcom,sdm845-dp
+          - qcom,sm8350-dp
+      - items:
+          - enum:
+              - qcom,sm8450-dp
+          - const: qcom,sm8350-dp

  reg:
    minItems: 4
--- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
@ -25,16 +25,16 @@ properties:
              - qcom,sc7280-dsi-ctrl
              - qcom,sdm660-dsi-ctrl
              - qcom,sdm845-dsi-ctrl
+              - qcom,sm6115-dsi-ctrl
              - qcom,sm8150-dsi-ctrl
              - qcom,sm8250-dsi-ctrl
              - qcom,sm8350-dsi-ctrl
              - qcom,sm8450-dsi-ctrl
              - qcom,sm8550-dsi-ctrl
          - const: qcom,mdss-dsi-ctrl
-      - items:
-          - enum:
-              - dsi-ctrl-6g-qcm2290
-          - const: qcom,mdss-dsi-ctrl
+      - enum:
+          - qcom,dsi-ctrl-6g-qcm2290
+          - qcom,mdss-dsi-ctrl # This should always come with an SoC-specific compatible
        deprecated: true

  reg:
@ -351,6 +351,7 @@ allOf:
          contains:
            enum:
              - qcom,sdm845-dsi-ctrl
+              - qcom,sm6115-dsi-ctrl
    then:
      properties:
        clocks:
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm6115-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm6115-mdss.yaml
@ -40,7 +40,13 @@ patternProperties:
    type: object
    properties:
      compatible:
-        const: qcom,dsi-ctrl-6g-qcm2290
+        oneOf:
+          - items:
+              - const: qcom,sm6115-dsi-ctrl
+              - const: qcom,mdss-dsi-ctrl
+          - description: Old binding, please don't use
+            deprecated: true
+            const: qcom,dsi-ctrl-6g-qcm2290

  "^phy@[0-9a-f]+$":
    type: object
@ -114,7 +120,7 @@ examples:
        };

        dsi@5e94000 {
-            compatible = "qcom,dsi-ctrl-6g-qcm2290";
+            compatible = "qcom,sm6115-dsi-ctrl", "qcom,mdss-dsi-ctrl";
            reg = <0x05e94000 0x400>;
            reg-names = "dsi_ctrl";

--- a/Documentation/devicetree/bindings/display/msm/qcom,sm8450-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8450-mdss.yaml
@ -54,7 +54,7 @@ patternProperties:
    type: object
    properties:
      compatible:
-        const: qcom,dsi-phy-5nm-8450
+        const: qcom,sm8450-dsi-phy-5nm

 required:
  - compatible
@ -254,7 +254,7 @@ examples:
        };

        dsi0_phy: phy@ae94400 {
-            compatible = "qcom,dsi-phy-5nm-8450";
+            compatible = "qcom,sm8450-dsi-phy-5nm";
            reg = <0x0ae94400 0x200>,
                  <0x0ae94600 0x280>,
                  <0x0ae94900 0x260>;
@ -325,7 +325,7 @@ examples:
        };

        dsi1_phy: phy@ae96400 {
-            compatible = "qcom,dsi-phy-5nm-8450";
+            compatible = "qcom,sm8450-dsi-phy-5nm";
            reg = <0x0ae96400 0x200>,
                  <0x0ae96600 0x280>,
                  <0x0ae96900 0x260>;
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm8550-dpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8550-dpu.yaml
@ -0,0 +1,133 @@
+# SPDX-License-Identifier: GPL-2.0-only or BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/msm/qcom,sm8550-dpu.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm SM8550 Display DPU
+
+maintainers:
+  - Neil Armstrong <neil.armstrong@linaro.org>
+
+$ref: /schemas/display/msm/dpu-common.yaml#
+
+properties:
+  compatible:
+    const: qcom,sm8550-dpu
+
+  reg:
+    items:
+      - description: Address offset and size for mdp register set
+      - description: Address offset and size for vbif register set
+
+  reg-names:
+    items:
+      - const: mdp
+      - const: vbif
+
+  clocks:
+    items:
+      - description: Display AHB
+      - description: Display hf axi
+      - description: Display MDSS ahb
+      - description: Display lut
+      - description: Display core
+      - description: Display vsync
+
+  clock-names:
+    items:
+      - const: bus
+      - const: nrt_bus
+      - const: iface
+      - const: lut
+      - const: core
+      - const: vsync
+
+required:
+  - compatible
+  - reg
+  - reg-names
+  - clocks
+  - clock-names
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/qcom,sm8550-dispcc.h>
+    #include <dt-bindings/clock/qcom,sm8550-gcc.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/power/qcom-rpmpd.h>
+
+    display-controller@ae01000 {
+        compatible = "qcom,sm8550-dpu";
+        reg = <0x0ae01000 0x8f000>,
+              <0x0aeb0000 0x2008>;
+        reg-names = "mdp", "vbif";
+
+        clocks = <&gcc GCC_DISP_AHB_CLK>,
+                <&gcc GCC_DISP_HF_AXI_CLK>,
+                <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>,
+                <&dispcc DISP_CC_MDSS_MDP_CLK>,
+                <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+        clock-names = "bus",
+                      "nrt_bus",
+                      "iface",
+                      "lut",
+                      "core",
+                      "vsync";
+
+        assigned-clocks = <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+        assigned-clock-rates = <19200000>;
+
+        operating-points-v2 = <&mdp_opp_table>;
+        power-domains = <&rpmhpd SM8550_MMCX>;
+
+        interrupt-parent = <&mdss>;
+        interrupts = <0>;
+
+        ports {
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            port@0 {
+                reg = <0>;
+                dpu_intf1_out: endpoint {
+                    remote-endpoint = <&dsi0_in>;
+                };
+            };
+
+            port@1 {
+                reg = <1>;
+                dpu_intf2_out: endpoint {
+                    remote-endpoint = <&dsi1_in>;
+                };
+            };
+        };
+
+        mdp_opp_table: opp-table {
+            compatible = "operating-points-v2";
+
+            opp-200000000 {
+                opp-hz = /bits/ 64 <200000000>;
+                required-opps = <&rpmhpd_opp_low_svs>;
+            };
+
+            opp-325000000 {
+                opp-hz = /bits/ 64 <325000000>;
+                required-opps = <&rpmhpd_opp_svs>;
+            };
+
+            opp-375000000 {
+                opp-hz = /bits/ 64 <375000000>;
+                required-opps = <&rpmhpd_opp_svs_l1>;
+            };
+
+            opp-514000000 {
+                opp-hz = /bits/ 64 <514000000>;
+                required-opps = <&rpmhpd_opp_nom>;
+            };
+        };
+    };
+...
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm8550-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8550-mdss.yaml
@ -0,0 +1,333 @@
+# SPDX-License-Identifier: GPL-2.0-only or BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/msm/qcom,sm8550-mdss.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm SM8550 Display MDSS
+
+maintainers:
+  - Neil Armstrong <neil.armstrong@linaro.org>
+
+description:
+  SM8550 MSM Mobile Display Subsystem(MDSS), which encapsulates sub-blocks like
+  DPU display controller, DSI and DP interfaces etc.
+
+$ref: /schemas/display/msm/mdss-common.yaml#
+
+properties:
+  compatible:
+    const: qcom,sm8550-mdss
+
+  clocks:
+    items:
+      - description: Display MDSS AHB
+      - description: Display AHB
+      - description: Display hf AXI
+      - description: Display core
+
+  iommus:
+    maxItems: 1
+
+  interconnects:
+    maxItems: 2
+
+  interconnect-names:
+    maxItems: 2
+
+patternProperties:
+  "^display-controller@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        const: qcom,sm8550-dpu
+
+  "^dsi@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        items:
+          - const: qcom,sm8550-dsi-ctrl
+          - const: qcom,mdss-dsi-ctrl
+
+  "^phy@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        const: qcom,sm8550-dsi-phy-4nm
+
+required:
+  - compatible
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/qcom,sm8550-dispcc.h>
+    #include <dt-bindings/clock/qcom,sm8550-gcc.h>
+    #include <dt-bindings/clock/qcom,rpmh.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/interconnect/qcom,sm8550-rpmh.h>
+    #include <dt-bindings/power/qcom-rpmpd.h>
+
+    display-subsystem@ae00000 {
+        compatible = "qcom,sm8550-mdss";
+        reg = <0x0ae00000 0x1000>;
+        reg-names = "mdss";
+
+        interconnects = <&mmss_noc MASTER_MDP 0 &gem_noc SLAVE_LLCC 0>,
+                        <&mc_virt MASTER_LLCC 0 &mc_virt SLAVE_EBI1 0>;
+        interconnect-names = "mdp0-mem", "mdp1-mem";
+
+        resets = <&dispcc DISP_CC_MDSS_CORE_BCR>;
+
+        power-domains = <&dispcc MDSS_GDSC>;
+
+        clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                 <&gcc GCC_DISP_AHB_CLK>,
+                 <&gcc GCC_DISP_HF_AXI_CLK>,
+                 <&dispcc DISP_CC_MDSS_MDP_CLK>;
+        clock-names = "iface", "bus", "nrt_bus", "core";
+
+        interrupts = <GIC_SPI 83 IRQ_TYPE_LEVEL_HIGH>;
+        interrupt-controller;
+        #interrupt-cells = <1>;
+
+        iommus = <&apps_smmu 0x1c00 0x2>;
+
+        #address-cells = <1>;
+        #size-cells = <1>;
+        ranges;
+
+        display-controller@ae01000 {
+            compatible = "qcom,sm8550-dpu";
+            reg = <0x0ae01000 0x8f000>,
+                  <0x0aeb0000 0x2008>;
+            reg-names = "mdp", "vbif";
+
+            clocks = <&gcc GCC_DISP_AHB_CLK>,
+                    <&gcc GCC_DISP_HF_AXI_CLK>,
+                    <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                    <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>,
+                    <&dispcc DISP_CC_MDSS_MDP_CLK>,
+                    <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+            clock-names = "bus",
+                          "nrt_bus",
+                          "iface",
+                          "lut",
+                          "core",
+                          "vsync";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+            assigned-clock-rates = <19200000>;
+
+            operating-points-v2 = <&mdp_opp_table>;
+            power-domains = <&rpmhpd SM8550_MMCX>;
+
+            interrupt-parent = <&mdss>;
+            interrupts = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dpu_intf1_out: endpoint {
+                        remote-endpoint = <&dsi0_in>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dpu_intf2_out: endpoint {
+                        remote-endpoint = <&dsi1_in>;
+                    };
+                };
+            };
+
+            mdp_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-200000000 {
+                    opp-hz = /bits/ 64 <200000000>;
+                    required-opps = <&rpmhpd_opp_low_svs>;
+                };
+
+                opp-325000000 {
+                    opp-hz = /bits/ 64 <325000000>;
+                    required-opps = <&rpmhpd_opp_svs>;
+                };
+
+                opp-375000000 {
+                    opp-hz = /bits/ 64 <375000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+
+                opp-514000000 {
+                    opp-hz = /bits/ 64 <514000000>;
+                    required-opps = <&rpmhpd_opp_nom>;
+                };
+            };
+        };
+
+        dsi@ae94000 {
+            compatible = "qcom,sm8550-dsi-ctrl", "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae94000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <4>;
+
+            clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK>,
+                     <&dispcc DISP_CC_MDSS_BYTE0_INTF_CLK>,
+                     <&dispcc DISP_CC_MDSS_PCLK0_CLK>,
+                     <&dispcc DISP_CC_MDSS_ESC0_CLK>,
+                     <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&gcc GCC_DISP_HF_AXI_CLK>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK_SRC>,
+                              <&dispcc DISP_CC_MDSS_PCLK0_CLK_SRC>;
+            assigned-clock-parents = <&dsi0_phy 0>, <&dsi0_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmhpd SM8550_MMCX>;
+
+            phys = <&dsi0_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dsi0_in: endpoint {
+                        remote-endpoint = <&dpu_intf1_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dsi0_out: endpoint {
+                    };
+                };
+            };
+
+            dsi_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-187500000 {
+                    opp-hz = /bits/ 64 <187500000>;
+                    required-opps = <&rpmhpd_opp_low_svs>;
+                };
+
+                opp-300000000 {
+                    opp-hz = /bits/ 64 <300000000>;
+                    required-opps = <&rpmhpd_opp_svs>;
+                };
+
+                opp-358000000 {
+                    opp-hz = /bits/ 64 <358000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+            };
+        };
+
+        dsi0_phy: phy@ae94400 {
+            compatible = "qcom,sm8550-dsi-phy-4nm";
+            reg = <0x0ae95000 0x200>,
+                  <0x0ae95200 0x280>,
+                  <0x0ae95500 0x400>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&rpmhcc RPMH_CXO_CLK>;
+            clock-names = "iface", "ref";
+        };
+
+        dsi@ae96000 {
+            compatible = "qcom,sm8550-dsi-ctrl", "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae96000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <5>;
+
+            clocks = <&dispcc DISP_CC_MDSS_BYTE1_CLK>,
+                     <&dispcc DISP_CC_MDSS_BYTE1_INTF_CLK>,
+                     <&dispcc DISP_CC_MDSS_PCLK1_CLK>,
+                     <&dispcc DISP_CC_MDSS_ESC1_CLK>,
+                     <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&gcc GCC_DISP_HF_AXI_CLK>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE1_CLK_SRC>,
+                              <&dispcc DISP_CC_MDSS_PCLK1_CLK_SRC>;
+            assigned-clock-parents = <&dsi1_phy 0>, <&dsi1_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmhpd SM8550_MMCX>;
+
+            phys = <&dsi1_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dsi1_in: endpoint {
+                        remote-endpoint = <&dpu_intf2_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dsi1_out: endpoint {
+                    };
+                };
+            };
+        };
+
+        dsi1_phy: phy@ae96400 {
+            compatible = "qcom,sm8550-dsi-phy-4nm";
+            reg = <0x0ae97000 0x200>,
+                  <0x0ae97200 0x280>,
+                  <0x0ae97500 0x400>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&rpmhcc RPMH_CXO_CLK>;
+            clock-names = "iface", "ref";
+        };
+    };
+...
--- a/2
+++ b/2
@ -6518,6 +6518,7 @@ L:	linux-arm-msm@vger.kernel.org
 L:	dri-devel@lists.freedesktop.org
 L:	freedreno@lists.freedesktop.org
 S:	Maintained
+B:	https://gitlab.freedesktop.org/drm/msm/-/issues
 T:	git https://gitlab.freedesktop.org/drm/msm.git
 F:	Documentation/devicetree/bindings/display/msm/
 F:	drivers/gpu/drm/msm/
@ -6935,6 +6936,7 @@ F:	drivers/phy/mediatek/phy-mtk-mipi*

 DRM DRIVERS FOR NVIDIA TEGRA
 M:	Thierry Reding <thierry.reding@gmail.com>
+M:	Mikko Perttunen <mperttunen@nvidia.com>
 L:	dri-devel@lists.freedesktop.org
 L:	linux-tegra@vger.kernel.org
 S:	Supported
--- a/drivers/accel/habanalabs/common/command_buffer.c
+++ b/drivers/accel/habanalabs/common/command_buffer.c
@ -45,20 +45,29 @@ static int cb_map_mem(struct hl_ctx *ctx, struct hl_cb *cb)
 	}

 	mutex_lock(&hdev->mmu_lock);
+
 	rc = hl_mmu_map_contiguous(ctx, cb->virtual_addr, cb->bus_address, cb->roundup_size);
 	if (rc) {
 		dev_err(hdev->dev, "Failed to map VA %#llx to CB\n", cb->virtual_addr);
-		goto err_va_umap;
+		goto err_va_pool_free;
 	}
+
 	rc = hl_mmu_invalidate_cache(hdev, false, MMU_OP_USERPTR | MMU_OP_SKIP_LOW_CACHE_INV);
+	if (rc)
+		goto err_mmu_unmap;
+
 	mutex_unlock(&hdev->mmu_lock);

 	cb->is_mmu_mapped = true;
-	return rc;

-err_va_umap:
+	return 0;
+
+err_mmu_unmap:
+	hl_mmu_unmap_contiguous(ctx, cb->virtual_addr, cb->roundup_size);
+err_va_pool_free:
 	mutex_unlock(&hdev->mmu_lock);
 	gen_pool_free(ctx->cb_va_pool, cb->virtual_addr, cb->roundup_size);
+
 	return rc;
 }

--- a/drivers/accel/habanalabs/common/decoder.c
+++ b/drivers/accel/habanalabs/common/decoder.c
@ -43,48 +43,46 @@ static void dec_print_abnrm_intr_source(struct hl_device *hdev, u32 irq_status)
 		intr_source[2], intr_source[3], intr_source[4], intr_source[5]);
 }

-static void dec_error_intr_work(struct hl_device *hdev, u32 base_addr, u32 core_id)
+static void dec_abnrm_intr_work(struct work_struct *work)
 {
+	struct hl_dec *dec = container_of(work, struct hl_dec, abnrm_intr_work);
+	struct hl_device *hdev = dec->hdev;
+	u32 irq_status, event_mask = 0;
 	bool reset_required = false;
-	u32 irq_status, event_mask;

-	irq_status = RREG32(base_addr + VCMD_IRQ_STATUS_OFFSET);
+	irq_status = RREG32(dec->base_addr + VCMD_IRQ_STATUS_OFFSET);

-	dev_err(hdev->dev, "Decoder abnormal interrupt %#x, core %d\n", irq_status, core_id);
+	dev_err(hdev->dev, "Decoder abnormal interrupt %#x, core %d\n", irq_status, dec->core_id);

 	dec_print_abnrm_intr_source(hdev, irq_status);

 	/* Clear the interrupt */
-	WREG32(base_addr + VCMD_IRQ_STATUS_OFFSET, irq_status);
+	WREG32(dec->base_addr + VCMD_IRQ_STATUS_OFFSET, irq_status);

 	/* Flush the interrupt clear */
-	RREG32(base_addr + VCMD_IRQ_STATUS_OFFSET);
+	RREG32(dec->base_addr + VCMD_IRQ_STATUS_OFFSET);

 	if (irq_status & VCMD_IRQ_STATUS_TIMEOUT_MASK) {
 		reset_required = true;
-		event_mask = HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
-	} else if (irq_status & VCMD_IRQ_STATUS_CMDERR_MASK) {
-		event_mask = HL_NOTIFIER_EVENT_UNDEFINED_OPCODE;
-	} else {
-		event_mask = HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
+		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 	}

+	if (irq_status & VCMD_IRQ_STATUS_CMDERR_MASK)
+		event_mask |= HL_NOTIFIER_EVENT_UNDEFINED_OPCODE;
+
+	if (irq_status & (VCMD_IRQ_STATUS_ENDCMD_MASK |
+				VCMD_IRQ_STATUS_BUSERR_MASK |
+				VCMD_IRQ_STATUS_ABORT_MASK))
+		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
+
 	if (reset_required) {
 		event_mask |= HL_NOTIFIER_EVENT_DEVICE_RESET;
 		hl_device_cond_reset(hdev, 0, event_mask);
-	} else {
+	} else if (event_mask) {
 		hl_notifier_event_send_all(hdev, event_mask);
 	}
 }

-static void dec_completion_abnrm(struct work_struct *work)
-{
-	struct hl_dec *dec = container_of(work, struct hl_dec, completion_abnrm_work);
-	struct hl_device *hdev = dec->hdev;
-
-	dec_error_intr_work(hdev, dec->base_addr, dec->core_id);
-}
-
 void hl_dec_fini(struct hl_device *hdev)
 {
 	kfree(hdev->dec);
@ -108,7 +106,7 @@ int hl_dec_init(struct hl_device *hdev)
 		dec = hdev->dec + j;

 		dec->hdev = hdev;
-		INIT_WORK(&dec->completion_abnrm_work, dec_completion_abnrm);
+		INIT_WORK(&dec->abnrm_intr_work, dec_abnrm_intr_work);
 		dec->core_id = j;
 		dec->base_addr = hdev->asic_funcs->get_dec_base_addr(hdev, j);
 		if (!dec->base_addr) {
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@ -1271,7 +1271,6 @@ int hl_device_resume(struct hl_device *hdev)
 	return 0;

 disable_device:
-	pci_clear_master(hdev->pdev);
 	pci_disable_device(hdev->pdev);

 	return rc;
@ -1381,6 +1380,34 @@ static void device_disable_open_processes(struct hl_device *hdev, bool control_d
 	mutex_unlock(fd_lock);
 }

+static void send_disable_pci_access(struct hl_device *hdev, u32 flags)
+{
+	/* If reset is due to heartbeat, device CPU is no responsive in
+	 * which case no point sending PCI disable message to it.
+	 */
+	if ((flags & HL_DRV_RESET_HARD) &&
+			!(flags & (HL_DRV_RESET_HEARTBEAT | HL_DRV_RESET_BYPASS_REQ_TO_FW))) {
+		/* Disable PCI access from device F/W so he won't send
+		 * us additional interrupts. We disable MSI/MSI-X at
+		 * the halt_engines function and we can't have the F/W
+		 * sending us interrupts after that. We need to disable
+		 * the access here because if the device is marked
+		 * disable, the message won't be send. Also, in case
+		 * of heartbeat, the device CPU is marked as disable
+		 * so this message won't be sent
+		 */
+		if (hl_fw_send_pci_access_msg(hdev, CPUCP_PACKET_DISABLE_PCI_ACCESS, 0x0)) {
+			dev_warn(hdev->dev, "Failed to disable FW's PCI access\n");
+			return;
+		}
+
+		/* verify that last EQs are handled before disabled is set */
+		if (hdev->cpu_queues_enable)
+			synchronize_irq(pci_irq_vector(hdev->pdev,
+					hdev->asic_prop.eq_interrupt_id));
+	}
+}
+
 static void handle_reset_trigger(struct hl_device *hdev, u32 flags)
 {
 	u32 cur_reset_trigger = HL_RESET_TRIGGER_DEFAULT;
@ -1419,28 +1446,6 @@ static void handle_reset_trigger(struct hl_device *hdev, u32 flags)
 	} else {
 		hdev->reset_info.reset_trigger_repeated = 1;
 	}
-
-	/* If reset is due to heartbeat, device CPU is no responsive in
-	 * which case no point sending PCI disable message to it.
-	 *
-	 * If F/W is performing the reset, no need to send it a message to disable
-	 * PCI access
-	 */
-	if ((flags & HL_DRV_RESET_HARD) &&
-			!(flags & (HL_DRV_RESET_HEARTBEAT | HL_DRV_RESET_BYPASS_REQ_TO_FW))) {
-		/* Disable PCI access from device F/W so he won't send
-		 * us additional interrupts. We disable MSI/MSI-X at
-		 * the halt_engines function and we can't have the F/W
-		 * sending us interrupts after that. We need to disable
-		 * the access here because if the device is marked
-		 * disable, the message won't be send. Also, in case
-		 * of heartbeat, the device CPU is marked as disable
-		 * so this message won't be sent
-		 */
-		if (hl_fw_send_pci_access_msg(hdev, CPUCP_PACKET_DISABLE_PCI_ACCESS, 0x0))
-			dev_warn(hdev->dev,
-				"Failed to disable FW's PCI access\n");
-	}
 }

 /*
@ -1561,6 +1566,7 @@ do_reset:

 escalate_reset_flow:
 		handle_reset_trigger(hdev, flags);
+		send_disable_pci_access(hdev, flags);

 		/* This also blocks future CS/VM/JOB completion operations */
 		hdev->disabled = true;
@ -1823,9 +1829,7 @@ kill_processes:
 			dev_info(hdev->dev, "Performing hard reset scheduled during compute reset\n");
 			flags = hdev->reset_info.hard_reset_schedule_flags;
 			hdev->reset_info.hard_reset_schedule_flags = 0;
-			hdev->disabled = true;
 			hard_reset = true;
-			handle_reset_trigger(hdev, flags);
 			goto escalate_reset_flow;
 		}
 	}
--- a/drivers/accel/habanalabs/common/firmware_if.c
+++ b/drivers/accel/habanalabs/common/firmware_if.c
@ -71,7 +71,7 @@ free_fw_ver:
 	return NULL;
 }

-static int extract_fw_sub_versions(struct hl_device *hdev, char *preboot_ver)
+static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver)
 {
 	char major[8], minor[8], *first_dot, *second_dot;
 	int rc;
@ -86,7 +86,7 @@ static int extract_fw_sub_versions(struct hl_device *hdev, char *preboot_ver)

 	if (rc) {
 		dev_err(hdev->dev, "Error %d parsing preboot major version\n", rc);
-		goto out;
+		return rc;
 	}

 	/* skip the first dot */
@ -102,9 +102,6 @@ static int extract_fw_sub_versions(struct hl_device *hdev, char *preboot_ver)

 	if (rc)
 		dev_err(hdev->dev, "Error %d parsing preboot minor version\n", rc);
-
-out:
-	kfree(preboot_ver);
 	return rc;
 }

@ -1263,7 +1260,7 @@ void hl_fw_ask_hard_reset_without_linux(struct hl_device *hdev)
 				COMMS_RST_DEV, 0, false,
 				hdev->fw_loader.cpu_timeout);
 		if (rc)
-			dev_warn(hdev->dev, "Failed sending COMMS_RST_DEV\n");
+			dev_err(hdev->dev, "Failed sending COMMS_RST_DEV\n");
 	} else {
 		WREG32(static_loader->kmd_msg_to_cpu_reg, KMD_MSG_RST_DEV);
 	}
@ -1281,10 +1278,10 @@ void hl_fw_ask_halt_machine_without_linux(struct hl_device *hdev)
 	/* Stop device CPU to make sure nothing bad happens */
 	if (hdev->asic_prop.dynamic_fw_load) {
 		rc = hl_fw_dynamic_send_protocol_cmd(hdev, &hdev->fw_loader,
-				COMMS_GOTO_WFE, 0, true,
+				COMMS_GOTO_WFE, 0, false,
 				hdev->fw_loader.cpu_timeout);
 		if (rc)
-			dev_warn(hdev->dev, "Failed sending COMMS_GOTO_WFE\n");
+			dev_err(hdev->dev, "Failed sending COMMS_GOTO_WFE\n");
 	} else {
 		WREG32(static_loader->kmd_msg_to_cpu_reg, KMD_MSG_GOTO_WFE);
 		msleep(static_loader->cpu_reset_wait_msec);
@ -2181,8 +2178,8 @@ static int hl_fw_dynamic_read_device_fw_version(struct hl_device *hdev,

 			dev_info(hdev->dev, "preboot version %s\n", preboot_ver);

-			/* This function takes care of freeing preboot_ver */
-			rc = extract_fw_sub_versions(hdev, preboot_ver);
+			rc = hl_get_preboot_major_minor(hdev, preboot_ver);
+			kfree(preboot_ver);
 			if (rc)
 				return rc;
 		}
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@ -662,7 +662,7 @@ struct hl_hints_range {
 * @user_interrupt_count: number of user interrupts.
 * @user_dec_intr_count: number of decoder interrupts exposed to user.
 * @tpc_interrupt_id: interrupt id for TPC to use in order to raise events towards the host.
- * @unexpected_user_error_interrupt_id: interrupt id used to indicate an unexpected user error.
+ * @eq_interrupt_id: interrupt id for EQ, uses to synchronize EQ interrupts in hard-reset.
 * @cache_line_size: device cache line size.
 * @server_type: Server type that the ASIC is currently installed in.
 *               The value is according to enum hl_server_type in uapi file.
@ -793,7 +793,7 @@ struct asic_fixed_properties {
 	u16				user_interrupt_count;
 	u16				user_dec_intr_count;
 	u16				tpc_interrupt_id;
-	u16				unexpected_user_error_interrupt_id;
+	u16				eq_interrupt_id;
 	u16				cache_line_size;
 	u16				server_type;
 	u8				completion_queues_count;
@ -1211,15 +1211,15 @@ struct hl_eq {
 /**
 * struct hl_dec - describes a decoder sw instance.
 * @hdev: pointer to the device structure.
- * @completion_abnrm_work: workqueue object to run when decoder generates an error interrupt
+ * @abnrm_intr_work: workqueue work item to run when decoder generates an error interrupt.
 * @core_id: ID of the decoder.
 * @base_addr: base address of the decoder.
 */
 struct hl_dec {
-	struct hl_device		*hdev;
-	struct work_struct		completion_abnrm_work;
-	u32				core_id;
-	u32				base_addr;
+	struct hl_device	*hdev;
+	struct work_struct	abnrm_intr_work;
+	u32			core_id;
+	u32			base_addr;
 };

 /**
--- a/drivers/accel/habanalabs/common/irq.c
+++ b/drivers/accel/habanalabs/common/irq.c
@ -415,8 +415,8 @@ irqreturn_t hl_irq_handler_eq(int irq, void *arg)
 	struct hl_eq_entry *eq_base;
 	struct hl_eqe_work *handle_eqe_work;
 	bool entry_ready;
-	u32 cur_eqe;
-	u16 cur_eqe_index;
+	u32 cur_eqe, ctl;
+	u16 cur_eqe_index, event_type;

 	eq_base = eq->kernel_address;

@ -449,7 +449,10 @@ irqreturn_t hl_irq_handler_eq(int irq, void *arg)
 		dma_rmb();

 		if (hdev->disabled && !hdev->reset_info.in_compute_reset) {
-			dev_warn(hdev->dev, "Device disabled but received an EQ event\n");
+			ctl = le32_to_cpu(eq_entry->hdr.ctl);
+			event_type = ((ctl & EQ_CTL_EVENT_TYPE_MASK) >> EQ_CTL_EVENT_TYPE_SHIFT);
+			dev_warn(hdev->dev,
+				"Device disabled but received an EQ event (%u)\n", event_type);
 			goto skip_irq;
 		}

@ -486,7 +489,7 @@ irqreturn_t hl_irq_handler_dec_abnrm(int irq, void *arg)
 {
 	struct hl_dec *dec = arg;

-	schedule_work(&dec->completion_abnrm_work);
+	schedule_work(&dec->abnrm_intr_work);

 	return IRQ_HANDLED;
 }
--- a/drivers/accel/habanalabs/common/memory.c
+++ b/drivers/accel/habanalabs/common/memory.c
@ -605,6 +605,7 @@ static u64 get_va_block(struct hl_device *hdev,
 	bool is_align_pow_2  = is_power_of_2(va_range->page_size);
 	bool is_hint_dram_addr = hl_is_dram_va(hdev, hint_addr);
 	bool force_hint = flags & HL_MEM_FORCE_HINT;
+	int rc;

 	if (is_align_pow_2)
 		align_mask = ~((u64)va_block_align - 1);
@ -722,9 +723,13 @@ static u64 get_va_block(struct hl_device *hdev,
 		kfree(new_va_block);
 	}

-	if (add_prev)
-		add_va_block_locked(hdev, &va_range->list, prev_start,
-				prev_end);
+	if (add_prev) {
+		rc = add_va_block_locked(hdev, &va_range->list, prev_start, prev_end);
+		if (rc) {
+			reserved_valid_start = 0;
+			goto out;
+		}
+	}

 	print_va_list_locked(hdev, &va_range->list);
 out:
--- a/drivers/accel/habanalabs/common/mmu/mmu.c
+++ b/drivers/accel/habanalabs/common/mmu/mmu.c
@ -679,7 +679,9 @@ int hl_mmu_invalidate_cache(struct hl_device *hdev, bool is_hard, u32 flags)

 	rc = hdev->asic_funcs->mmu_invalidate_cache(hdev, is_hard, flags);
 	if (rc)
-		dev_err_ratelimited(hdev->dev, "MMU cache invalidation failed\n");
+		dev_err_ratelimited(hdev->dev,
+				"%s cache invalidation failed, rc=%d\n",
+				flags == VM_TYPE_USERPTR ? "PMMU" : "HMMU", rc);

 	return rc;
 }
@ -692,7 +694,9 @@ int hl_mmu_invalidate_cache_range(struct hl_device *hdev, bool is_hard,
 	rc = hdev->asic_funcs->mmu_invalidate_cache_range(hdev, is_hard, flags,
 								asid, va, size);
 	if (rc)
-		dev_err_ratelimited(hdev->dev, "MMU cache range invalidation failed\n");
+		dev_err_ratelimited(hdev->dev,
+				"%s cache range invalidation failed: va=%#llx, size=%llu, rc=%d",
+				flags == VM_TYPE_USERPTR ? "PMMU" : "HMMU", va, size, rc);

 	return rc;
 }
--- a/drivers/accel/habanalabs/common/pci/pci.c
+++ b/drivers/accel/habanalabs/common/pci/pci.c
@ -420,7 +420,6 @@ int hl_pci_init(struct hl_device *hdev)
 unmap_pci_bars:
 	hl_pci_bars_unmap(hdev);
 disable_device:
-	pci_clear_master(pdev);
 	pci_disable_device(pdev);

 	return rc;
@ -436,6 +435,5 @@ void hl_pci_fini(struct hl_device *hdev)
 {
 	hl_pci_bars_unmap(hdev);

-	pci_clear_master(hdev->pdev);
 	pci_disable_device(hdev->pdev);
 }
--- a/drivers/accel/habanalabs/common/sysfs.c
+++ b/drivers/accel/habanalabs/common/sysfs.c
@ -497,10 +497,14 @@ int hl_sysfs_init(struct hl_device *hdev)
 	if (rc) {
 		dev_err(hdev->dev,
 			"Failed to add groups to device, error %d\n", rc);
-		return rc;
+		goto remove_groups;
 	}

 	return 0;
+
+remove_groups:
+	device_remove_groups(hdev->dev, hl_dev_attr_groups);
+	return rc;
 }

 void hl_sysfs_fini(struct hl_device *hdev)
--- a/drivers/accel/habanalabs/gaudi/gaudi.c
+++ b/drivers/accel/habanalabs/gaudi/gaudi.c
@ -682,6 +682,9 @@ static int gaudi_set_fixed_properties(struct hl_device *hdev)
 	prop->first_available_user_interrupt = USHRT_MAX;
 	prop->tpc_interrupt_id = USHRT_MAX;

+	/* single msi */
+	prop->eq_interrupt_id = 0;
+
 	for (i = 0 ; i < HL_MAX_DCORES ; i++)
 		prop->first_available_cq[i] = USHRT_MAX;

@ -2017,38 +2020,6 @@ static int gaudi_enable_msi_single(struct hl_device *hdev)
 	return rc;
 }

-static int gaudi_enable_msi_multi(struct hl_device *hdev)
-{
-	int cq_cnt = hdev->asic_prop.completion_queues_count;
-	int rc, i, irq_cnt_init, irq;
-
-	for (i = 0, irq_cnt_init = 0 ; i < cq_cnt ; i++, irq_cnt_init++) {
-		irq = gaudi_pci_irq_vector(hdev, i, false);
-		rc = request_irq(irq, hl_irq_handler_cq, 0, gaudi_irq_name[i],
-				&hdev->completion_queue[i]);
-		if (rc) {
-			dev_err(hdev->dev, "Failed to request IRQ %d", irq);
-			goto free_irqs;
-		}
-	}
-
-	irq = gaudi_pci_irq_vector(hdev, GAUDI_EVENT_QUEUE_MSI_IDX, true);
-	rc = request_irq(irq, hl_irq_handler_eq, 0, gaudi_irq_name[cq_cnt],
-				&hdev->event_queue);
-	if (rc) {
-		dev_err(hdev->dev, "Failed to request IRQ %d", irq);
-		goto free_irqs;
-	}
-
-	return 0;
-
-free_irqs:
-	for (i = 0 ; i < irq_cnt_init ; i++)
-		free_irq(gaudi_pci_irq_vector(hdev, i, false),
-				&hdev->completion_queue[i]);
-	return rc;
-}
-
 static int gaudi_enable_msi(struct hl_device *hdev)
 {
 	struct gaudi_device *gaudi = hdev->asic_specific;
@ -2063,14 +2034,7 @@ static int gaudi_enable_msi(struct hl_device *hdev)
 		return rc;
 	}

-	if (rc < NUMBER_OF_INTERRUPTS) {
-		gaudi->multi_msi_mode = false;
-		rc = gaudi_enable_msi_single(hdev);
-	} else {
-		gaudi->multi_msi_mode = true;
-		rc = gaudi_enable_msi_multi(hdev);
-	}
-
+	rc = gaudi_enable_msi_single(hdev);
 	if (rc)
 		goto free_pci_irq_vectors;

@ -2086,47 +2050,23 @@ free_pci_irq_vectors:
 static void gaudi_sync_irqs(struct hl_device *hdev)
 {
 	struct gaudi_device *gaudi = hdev->asic_specific;
-	int i, cq_cnt = hdev->asic_prop.completion_queues_count;

 	if (!(gaudi->hw_cap_initialized & HW_CAP_MSI))
 		return;

 	/* Wait for all pending IRQs to be finished */
-	if (gaudi->multi_msi_mode) {
-		for (i = 0 ; i < cq_cnt ; i++)
-			synchronize_irq(gaudi_pci_irq_vector(hdev, i, false));
-
-		synchronize_irq(gaudi_pci_irq_vector(hdev,
-						GAUDI_EVENT_QUEUE_MSI_IDX,
-						true));
-	} else {
-		synchronize_irq(gaudi_pci_irq_vector(hdev, 0, false));
-	}
+	synchronize_irq(gaudi_pci_irq_vector(hdev, 0, false));
 }

 static void gaudi_disable_msi(struct hl_device *hdev)
 {
 	struct gaudi_device *gaudi = hdev->asic_specific;
-	int i, irq, cq_cnt = hdev->asic_prop.completion_queues_count;

 	if (!(gaudi->hw_cap_initialized & HW_CAP_MSI))
 		return;

 	gaudi_sync_irqs(hdev);
-
-	if (gaudi->multi_msi_mode) {
-		irq = gaudi_pci_irq_vector(hdev, GAUDI_EVENT_QUEUE_MSI_IDX,
-						true);
-		free_irq(irq, &hdev->event_queue);
-
-		for (i = 0 ; i < cq_cnt ; i++) {
-			irq = gaudi_pci_irq_vector(hdev, i, false);
-			free_irq(irq, &hdev->completion_queue[i]);
-		}
-	} else {
-		free_irq(gaudi_pci_irq_vector(hdev, 0, false), hdev);
-	}
-
+	free_irq(gaudi_pci_irq_vector(hdev, 0, false), hdev);
 	pci_free_irq_vectors(hdev->pdev);

 	gaudi->hw_cap_initialized &= ~HW_CAP_MSI;
@ -3921,11 +3861,7 @@ static int gaudi_init_cpu_queues(struct hl_device *hdev, u32 cpu_timeout)

 	WREG32(mmCPU_IF_PF_PQ_PI, 0);

-	if (gaudi->multi_msi_mode)
-		WREG32(mmCPU_IF_QUEUE_INIT, PQ_INIT_STATUS_READY_FOR_CP);
-	else
-		WREG32(mmCPU_IF_QUEUE_INIT,
-			PQ_INIT_STATUS_READY_FOR_CP_SINGLE_MSI);
+	WREG32(mmCPU_IF_QUEUE_INIT, PQ_INIT_STATUS_READY_FOR_CP_SINGLE_MSI);

 	irq_handler_offset = prop->gic_interrupts_enable ?
 			mmGIC_DISTRIBUTOR__5_GICD_SETSPI_NSR :
@ -5602,7 +5538,6 @@ static void gaudi_add_end_of_cb_packets(struct hl_device *hdev, void *kernel_add
 				u32 len, u32 original_len, u64 cq_addr, u32 cq_val,
 				u32 msi_vec, bool eb)
 {
-	struct gaudi_device *gaudi = hdev->asic_specific;
 	struct packet_msg_prot *cq_pkt;
 	struct packet_nop *cq_padding;
 	u64 msi_addr;
@ -5632,12 +5567,7 @@ static void gaudi_add_end_of_cb_packets(struct hl_device *hdev, void *kernel_add
 	tmp |= FIELD_PREP(GAUDI_PKT_CTL_MB_MASK, 1);
 	cq_pkt->ctl = cpu_to_le32(tmp);
 	cq_pkt->value = cpu_to_le32(1);
-
-	if (gaudi->multi_msi_mode)
-		msi_addr = mmPCIE_MSI_INTR_0 + msi_vec * 4;
-	else
-		msi_addr = mmPCIE_CORE_MSI_REQ;
-
+	msi_addr = hdev->pdev ? mmPCIE_CORE_MSI_REQ : mmPCIE_MSI_INTR_0 + msi_vec * 4;
 	cq_pkt->addr = cpu_to_le64(CFG_BASE + msi_addr);
 }

--- a/drivers/accel/habanalabs/gaudi/gaudiP.h
+++ b/drivers/accel/habanalabs/gaudi/gaudiP.h
@ -28,20 +28,8 @@
 #define NUMBER_OF_COLLECTIVE_QUEUES	12
 #define NUMBER_OF_SOBS_IN_GRP		11

-/*
- * Number of MSI interrupts IDS:
- * Each completion queue has 1 ID
- * The event queue has 1 ID
- */
-#define NUMBER_OF_INTERRUPTS		(NUMBER_OF_CMPLT_QUEUES + \
-						NUMBER_OF_CPU_HW_QUEUES)
-
 #define GAUDI_STREAM_MASTER_ARR_SIZE	8

-#if (NUMBER_OF_INTERRUPTS > GAUDI_MSI_ENTRIES)
-#error "Number of MSI interrupts must be smaller or equal to GAUDI_MSI_ENTRIES"
-#endif
-
 #define CORESIGHT_TIMEOUT_USEC		100000		/* 100 ms */

 #define GAUDI_MAX_CLK_FREQ		2200000000ull	/* 2200 MHz */
@ -324,8 +312,6 @@ struct gaudi_internal_qman_info {
 *                      signal we can use this engine in later code paths.
 *                      Each bit is cleared upon reset of its corresponding H/W
 *                      engine.
- * @multi_msi_mode: whether we are working in multi MSI single MSI mode.
- *                  Multi MSI is possible only with IOMMU enabled.
 * @mmu_cache_inv_pi: PI for MMU cache invalidation flow. The H/W expects an
 *                    8-bit value so use u8.
 */
@ -345,7 +331,6 @@ struct gaudi_device {
 	u32				events_stat[GAUDI_EVENT_SIZE];
 	u32				events_stat_aggregate[GAUDI_EVENT_SIZE];
 	u32				hw_cap_initialized;
-	u8				multi_msi_mode;
 	u8				mmu_cache_inv_pi;
 };

--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@ -2112,6 +2112,7 @@ static bool gaudi2_get_mme_idle_status(struct hl_device *hdev, u64 *mask_arr, u8
 static bool gaudi2_get_edma_idle_status(struct hl_device *hdev, u64 *mask_arr, u8 mask_len,
 		struct engines_data *e);
 static u64 gaudi2_mmu_scramble_addr(struct hl_device *hdev, u64 raw_addr);
+static u64 gaudi2_mmu_descramble_addr(struct hl_device *hdev, u64 scrambled_addr);

 static void gaudi2_init_scrambler_hbm(struct hl_device *hdev)
 {
@ -2438,7 +2439,7 @@ static int gaudi2_set_fixed_properties(struct hl_device *hdev)

 	prop->first_available_user_interrupt = GAUDI2_IRQ_NUM_USER_FIRST;
 	prop->tpc_interrupt_id = GAUDI2_IRQ_NUM_TPC_ASSERT;
-	prop->unexpected_user_error_interrupt_id = GAUDI2_IRQ_NUM_UNEXPECTED_ERROR;
+	prop->eq_interrupt_id = GAUDI2_IRQ_NUM_EVENT_QUEUE;

 	prop->first_available_cq[0] = GAUDI2_RESERVED_CQ_NUMBER;

@ -2887,6 +2888,10 @@ static int gaudi2_cpucp_info_get(struct hl_device *hdev)
 	hdev->tpc_binning = le64_to_cpu(prop->cpucp_info.tpc_binning_mask);
 	hdev->decoder_binning = lower_32_bits(le64_to_cpu(prop->cpucp_info.decoder_binning_mask));

+	dev_dbg(hdev->dev, "Read binning masks: tpc: 0x%llx, dram: 0x%llx, edma: 0x%x, dec: 0x%x\n",
+			hdev->tpc_binning, hdev->dram_binning, hdev->edma_binning,
+			hdev->decoder_binning);
+
 	/*
 	 * at this point the DRAM parameters need to be updated according to data obtained
 	 * from the FW
@ -3345,7 +3350,7 @@ static void gaudi2_user_interrupt_setup(struct hl_device *hdev)
 	/* Initialize TPC interrupt */
 	HL_USR_INTR_STRUCT_INIT(hdev->tpc_interrupt, hdev, 0, HL_USR_INTERRUPT_TPC);

-	/* Initialize general purpose interrupt */
+	/* Initialize unexpected error interrupt */
 	HL_USR_INTR_STRUCT_INIT(hdev->unexpected_error_interrupt, hdev, 0,
 						HL_USR_INTERRUPT_UNEXPECTED);

@ -3475,6 +3480,48 @@ static int gaudi2_special_blocks_iterator_config(struct hl_device *hdev)
 	return gaudi2_special_blocks_config(hdev);
 }

+static void gaudi2_test_queues_msgs_free(struct hl_device *hdev)
+{
+	struct gaudi2_device *gaudi2 = hdev->asic_specific;
+	struct gaudi2_queues_test_info *msg_info = gaudi2->queues_test_info;
+	int i;
+
+	for (i = 0 ; i < GAUDI2_NUM_TESTED_QS ; i++) {
+		/* bail-out if this is an allocation failure point */
+		if (!msg_info[i].kern_addr)
+			break;
+
+		hl_asic_dma_pool_free(hdev, msg_info[i].kern_addr, msg_info[i].dma_addr);
+		msg_info[i].kern_addr = NULL;
+	}
+}
+
+static int gaudi2_test_queues_msgs_alloc(struct hl_device *hdev)
+{
+	struct gaudi2_device *gaudi2 = hdev->asic_specific;
+	struct gaudi2_queues_test_info *msg_info = gaudi2->queues_test_info;
+	int i, rc;
+
+	/* allocate a message-short buf for each Q we intend to test */
+	for (i = 0 ; i < GAUDI2_NUM_TESTED_QS ; i++) {
+		msg_info[i].kern_addr =
+			(void *)hl_asic_dma_pool_zalloc(hdev, sizeof(struct packet_msg_short),
+							GFP_KERNEL, &msg_info[i].dma_addr);
+		if (!msg_info[i].kern_addr) {
+			dev_err(hdev->dev,
+				"Failed to allocate dma memory for H/W queue %d testing\n", i);
+			rc = -ENOMEM;
+			goto err_exit;
+		}
+	}
+
+	return 0;
+
+err_exit:
+	gaudi2_test_queues_msgs_free(hdev);
+	return rc;
+}
+
 static int gaudi2_sw_init(struct hl_device *hdev)
 {
 	struct asic_fixed_properties *prop = &hdev->asic_prop;
@ -3574,8 +3621,14 @@ static int gaudi2_sw_init(struct hl_device *hdev)
 	if (rc)
 		goto free_scratchpad_mem;

+	rc = gaudi2_test_queues_msgs_alloc(hdev);
+	if (rc)
+		goto special_blocks_free;
+
 	return 0;

+special_blocks_free:
+	gaudi2_special_blocks_iterator_free(hdev);
 free_scratchpad_mem:
 	hl_asic_dma_pool_free(hdev, gaudi2->scratchpad_kernel_address,
 				gaudi2->scratchpad_bus_address);
@ -3598,6 +3651,8 @@ static int gaudi2_sw_fini(struct hl_device *hdev)
 	struct asic_fixed_properties *prop = &hdev->asic_prop;
 	struct gaudi2_device *gaudi2 = hdev->asic_specific;

+	gaudi2_test_queues_msgs_free(hdev);
+
 	gaudi2_special_blocks_iterator_free(hdev);

 	hl_cpu_accessible_dma_pool_free(hdev, prop->pmmu.page_size, gaudi2->virt_msix_db_cpu_addr);
@ -4009,7 +4064,7 @@ static const char *gaudi2_irq_name(u16 irq_number)
 	case GAUDI2_IRQ_NUM_TPC_ASSERT:
 		return "gaudi2 tpc assert";
 	case GAUDI2_IRQ_NUM_UNEXPECTED_ERROR:
-		return "gaudi2 tpc assert";
+		return "gaudi2 unexpected error";
 	case GAUDI2_IRQ_NUM_USER_FIRST ... GAUDI2_IRQ_NUM_USER_LAST:
 		return "gaudi2 user completion";
 	default:
@ -6792,29 +6847,30 @@ static void gaudi2_qman_set_test_mode(struct hl_device *hdev, u32 hw_queue_id, b
 	}
 }

-static int gaudi2_test_queue(struct hl_device *hdev, u32 hw_queue_id)
+static inline u32 gaudi2_test_queue_hw_queue_id_to_sob_id(struct hl_device *hdev, u32 hw_queue_id)
 {
-	u32 sob_offset = hdev->asic_prop.first_available_user_sob[0] * 4;
+	return hdev->asic_prop.first_available_user_sob[0] +
+				hw_queue_id - GAUDI2_QUEUE_ID_PDMA_0_0;
+}
+
+static void gaudi2_test_queue_clear(struct hl_device *hdev, u32 hw_queue_id)
+{
+	u32 sob_offset = gaudi2_test_queue_hw_queue_id_to_sob_id(hdev, hw_queue_id) * 4;
 	u32 sob_addr = mmDCORE0_SYNC_MNGR_OBJS_SOB_OBJ_0 + sob_offset;
-	u32 timeout_usec, tmp, sob_base = 1, sob_val = 0x5a5a;
-	struct packet_msg_short *msg_short_pkt;
-	dma_addr_t pkt_dma_addr;
-	size_t pkt_size;
+
+	/* Reset the SOB value */
+	WREG32(sob_addr, 0);
+}
+
+static int gaudi2_test_queue_send_msg_short(struct hl_device *hdev, u32 hw_queue_id, u32 sob_val,
+					    struct gaudi2_queues_test_info *msg_info)
+{
+	u32 sob_offset =  gaudi2_test_queue_hw_queue_id_to_sob_id(hdev, hw_queue_id) * 4;
+	u32 tmp, sob_base = 1;
+	struct packet_msg_short *msg_short_pkt = msg_info->kern_addr;
+	size_t pkt_size = sizeof(struct packet_msg_short);
 	int rc;

-	if (hdev->pldm)
-		timeout_usec = GAUDI2_PLDM_TEST_QUEUE_WAIT_USEC;
-	else
-		timeout_usec = GAUDI2_TEST_QUEUE_WAIT_USEC;
-
-	pkt_size = sizeof(*msg_short_pkt);
-	msg_short_pkt = hl_asic_dma_pool_zalloc(hdev, pkt_size, GFP_KERNEL, &pkt_dma_addr);
-	if (!msg_short_pkt) {
-		dev_err(hdev->dev, "Failed to allocate packet for H/W queue %d testing\n",
-			hw_queue_id);
-		return -ENOMEM;
-	}
-
 	tmp = (PACKET_MSG_SHORT << GAUDI2_PKT_CTL_OPCODE_SHIFT) |
 		(1 << GAUDI2_PKT_CTL_EB_SHIFT) |
 		(1 << GAUDI2_PKT_CTL_MB_SHIFT) |
@ -6824,15 +6880,25 @@ static int gaudi2_test_queue(struct hl_device *hdev, u32 hw_queue_id)
 	msg_short_pkt->value = cpu_to_le32(sob_val);
 	msg_short_pkt->ctl = cpu_to_le32(tmp);

-	/* Reset the SOB value */
-	WREG32(sob_addr, 0);
+	rc = hl_hw_queue_send_cb_no_cmpl(hdev, hw_queue_id, pkt_size, msg_info->dma_addr);
+	if (rc)
+		dev_err(hdev->dev,
+			"Failed to send msg_short packet to H/W queue %d\n", hw_queue_id);

-	rc = hl_hw_queue_send_cb_no_cmpl(hdev, hw_queue_id, pkt_size, pkt_dma_addr);
-	if (rc) {
-		dev_err(hdev->dev, "Failed to send msg_short packet to H/W queue %d\n",
-			hw_queue_id);
-		goto free_pkt;
-	}
+	return rc;
+}
+
+static int gaudi2_test_queue_wait_completion(struct hl_device *hdev, u32 hw_queue_id, u32 sob_val)
+{
+	u32 sob_offset = gaudi2_test_queue_hw_queue_id_to_sob_id(hdev, hw_queue_id) * 4;
+	u32 sob_addr = mmDCORE0_SYNC_MNGR_OBJS_SOB_OBJ_0 + sob_offset;
+	u32 timeout_usec, tmp;
+	int rc;
+
+	if (hdev->pldm)
+		timeout_usec = GAUDI2_PLDM_TEST_QUEUE_WAIT_USEC;
+	else
+		timeout_usec = GAUDI2_TEST_QUEUE_WAIT_USEC;

 	rc = hl_poll_timeout(
 			hdev,
@ -6848,11 +6914,6 @@ static int gaudi2_test_queue(struct hl_device *hdev, u32 hw_queue_id)
 		rc = -EIO;
 	}

-	/* Reset the SOB value */
-	WREG32(sob_addr, 0);
-
-free_pkt:
-	hl_asic_dma_pool_free(hdev, (void *) msg_short_pkt, pkt_dma_addr);
 	return rc;
 }

@ -6872,30 +6933,44 @@ static int gaudi2_test_cpu_queue(struct hl_device *hdev)

 static int gaudi2_test_queues(struct hl_device *hdev)
 {
-	int i, rc, ret_val = 0;
+	struct gaudi2_device *gaudi2 = hdev->asic_specific;
+	struct gaudi2_queues_test_info *msg_info;
+	u32 sob_val = 0x5a5a;
+	int i, rc;

+	/* send test message on all enabled Qs */
 	for (i = GAUDI2_QUEUE_ID_PDMA_0_0 ; i < GAUDI2_QUEUE_ID_CPU_PQ; i++) {
 		if (!gaudi2_is_queue_enabled(hdev, i))
 			continue;

+		msg_info = &gaudi2->queues_test_info[i - GAUDI2_QUEUE_ID_PDMA_0_0];
 		gaudi2_qman_set_test_mode(hdev, i, true);
-		rc = gaudi2_test_queue(hdev, i);
-		gaudi2_qman_set_test_mode(hdev, i, false);
-
-		if (rc) {
-			ret_val = -EINVAL;
+		gaudi2_test_queue_clear(hdev, i);
+		rc = gaudi2_test_queue_send_msg_short(hdev, i, sob_val, msg_info);
+		if (rc)
 			goto done;
-		}
 	}

 	rc = gaudi2_test_cpu_queue(hdev);
-	if (rc) {
-		ret_val = -EINVAL;
+	if (rc)
 		goto done;
+
+	/* verify that all messages were processed */
+	for (i = GAUDI2_QUEUE_ID_PDMA_0_0 ; i < GAUDI2_QUEUE_ID_CPU_PQ; i++) {
+		if (!gaudi2_is_queue_enabled(hdev, i))
+			continue;
+
+		rc = gaudi2_test_queue_wait_completion(hdev, i, sob_val);
+		if (rc)
+			/* chip is not usable, no need for cleanups, just bail-out with error */
+			goto done;
+
+		gaudi2_test_queue_clear(hdev, i);
+		gaudi2_qman_set_test_mode(hdev, i, false);
 	}

 done:
-	return ret_val;
+	return rc;
 }

 static int gaudi2_compute_reset_late_init(struct hl_device *hdev)
@ -8485,23 +8560,28 @@ static int gaudi2_handle_qman_err(struct hl_device *hdev, u16 event_type, u64 *e

 static int gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev, u16 event_type)
 {
-	u32 i, sts_val, sts_clr_val = 0, error_count = 0;
+	u32 i, sts_val, sts_clr_val, error_count = 0, arc_farm;

-	sts_val = RREG32(mmARC_FARM_ARC0_AUX_ARC_SEI_INTR_STS);
+	for (arc_farm = 0 ; arc_farm < NUM_OF_ARC_FARMS_ARC ; arc_farm++) {
+		sts_clr_val = 0;
+		sts_val = RREG32(mmARC_FARM_ARC0_AUX_ARC_SEI_INTR_STS +
+				(arc_farm * ARC_FARM_OFFSET));

-	for (i = 0 ; i < GAUDI2_NUM_OF_ARC_SEI_ERR_CAUSE ; i++) {
-		if (sts_val & BIT(i)) {
-			gaudi2_print_event(hdev, event_type, true,
-				"err cause: %s", gaudi2_arc_sei_error_cause[i]);
-			sts_clr_val |= BIT(i);
-			error_count++;
+		for (i = 0 ; i < GAUDI2_NUM_OF_ARC_SEI_ERR_CAUSE ; i++) {
+			if (sts_val & BIT(i)) {
+				gaudi2_print_event(hdev, event_type, true,
+						"ARC FARM ARC %u err cause: %s",
+						arc_farm, gaudi2_arc_sei_error_cause[i]);
+				sts_clr_val |= BIT(i);
+				error_count++;
+			}
 		}
+		WREG32(mmARC_FARM_ARC0_AUX_ARC_SEI_INTR_CLR + (arc_farm * ARC_FARM_OFFSET),
+				sts_clr_val);
 	}

 	hl_check_for_glbl_errors(hdev);

-	WREG32(mmARC_FARM_ARC0_AUX_ARC_SEI_INTR_CLR, sts_clr_val);
-
 	return error_count;
 }

@ -8844,7 +8924,7 @@ static int gaudi2_handle_hif_fatal(struct hl_device *hdev, u16 event_type, u64 i
 static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool is_pmmu,
 					u64 *event_mask)
 {
-	u32 valid, val, axid_l, axid_h;
+	u32 valid, val;
 	u64 addr;

 	valid = RREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_PAGE_ERROR_VALID));
@ -8857,11 +8937,11 @@ static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool
 	addr <<= 32;
 	addr |= RREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_PAGE_ERROR_CAPTURE_VA));

-	axid_l = RREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_PAGE_FAULT_ID_LSB));
-	axid_h = RREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_PAGE_FAULT_ID_MSB));
+	if (!is_pmmu)
+		addr = gaudi2_mmu_descramble_addr(hdev, addr);

-	dev_err_ratelimited(hdev->dev, "%s page fault on va 0x%llx, transaction id 0x%llX\n",
-				is_pmmu ? "PMMU" : "HMMU", addr, ((u64)axid_h << 32) + axid_l);
+	dev_err_ratelimited(hdev->dev, "%s page fault on va 0x%llx\n",
+				is_pmmu ? "PMMU" : "HMMU", addr);
 	hl_handle_page_fault(hdev, addr, 0, is_pmmu, event_mask);

 	WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_PAGE_ERROR_VALID), 0);
@ -8882,9 +8962,12 @@ static void gaudi2_handle_access_error(struct hl_device *hdev, u64 mmu_base, boo
 	addr <<= 32;
 	addr |= RREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_ERROR_CAPTURE_VA));

+	if (!is_pmmu)
+		addr = gaudi2_mmu_descramble_addr(hdev, addr);
+
 	dev_err_ratelimited(hdev->dev, "%s access error on va 0x%llx\n",
 				is_pmmu ? "PMMU" : "HMMU", addr);
-	WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_ERROR_CAPTURE), 0);
+	WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_PAGE_ERROR_VALID), 0);
 }

 static int gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, u16 event_type,
@ -8976,46 +9059,110 @@ static int gaudi2_handle_sm_err(struct hl_device *hdev, u16 event_type, u8 sm_in
 	return error_count;
 }

+static u64 get_hmmu_base(u16 event_type)
+{
+	u8 dcore, index_in_dcore;
+
+	switch (event_type) {
+	case GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU0_SPI_BASE ... GAUDI2_EVENT_HMMU0_SECURITY_ERROR:
+		dcore = 0;
+		index_in_dcore = 0;
+	break;
+	case GAUDI2_EVENT_HMMU_1_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU1_SPI_BASE ... GAUDI2_EVENT_HMMU1_SECURITY_ERROR:
+		dcore = 1;
+		index_in_dcore = 0;
+	break;
+	case GAUDI2_EVENT_HMMU_2_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU2_SPI_BASE ... GAUDI2_EVENT_HMMU2_SECURITY_ERROR:
+		dcore = 0;
+		index_in_dcore = 1;
+	break;
+	case GAUDI2_EVENT_HMMU_3_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU3_SPI_BASE ... GAUDI2_EVENT_HMMU3_SECURITY_ERROR:
+		dcore = 1;
+		index_in_dcore = 1;
+	break;
+	case GAUDI2_EVENT_HMMU_4_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU4_SPI_BASE ... GAUDI2_EVENT_HMMU4_SECURITY_ERROR:
+		dcore = 3;
+		index_in_dcore = 2;
+	break;
+	case GAUDI2_EVENT_HMMU_5_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU5_SPI_BASE ... GAUDI2_EVENT_HMMU5_SECURITY_ERROR:
+		dcore = 2;
+		index_in_dcore = 2;
+	break;
+	case GAUDI2_EVENT_HMMU_6_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU6_SPI_BASE ... GAUDI2_EVENT_HMMU6_SECURITY_ERROR:
+		dcore = 3;
+		index_in_dcore = 3;
+	break;
+	case GAUDI2_EVENT_HMMU_7_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU7_SPI_BASE ... GAUDI2_EVENT_HMMU7_SECURITY_ERROR:
+		dcore = 2;
+		index_in_dcore = 3;
+	break;
+	case GAUDI2_EVENT_HMMU_8_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU8_SPI_BASE ... GAUDI2_EVENT_HMMU8_SECURITY_ERROR:
+		dcore = 0;
+		index_in_dcore = 2;
+	break;
+	case GAUDI2_EVENT_HMMU_9_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU9_SPI_BASE ... GAUDI2_EVENT_HMMU9_SECURITY_ERROR:
+		dcore = 1;
+		index_in_dcore = 2;
+	break;
+	case GAUDI2_EVENT_HMMU_10_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU10_SPI_BASE ... GAUDI2_EVENT_HMMU10_SECURITY_ERROR:
+		dcore = 0;
+		index_in_dcore = 3;
+	break;
+	case GAUDI2_EVENT_HMMU_11_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU11_SPI_BASE ... GAUDI2_EVENT_HMMU11_SECURITY_ERROR:
+		dcore = 1;
+		index_in_dcore = 3;
+	break;
+	case GAUDI2_EVENT_HMMU_12_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU12_SPI_BASE ... GAUDI2_EVENT_HMMU12_SECURITY_ERROR:
+		dcore = 3;
+		index_in_dcore = 0;
+	break;
+	case GAUDI2_EVENT_HMMU_13_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU13_SPI_BASE ... GAUDI2_EVENT_HMMU13_SECURITY_ERROR:
+		dcore = 2;
+		index_in_dcore = 0;
+	break;
+	case GAUDI2_EVENT_HMMU_14_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU14_SPI_BASE ... GAUDI2_EVENT_HMMU14_SECURITY_ERROR:
+		dcore = 3;
+		index_in_dcore = 1;
+	break;
+	case GAUDI2_EVENT_HMMU_15_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU15_SPI_BASE ... GAUDI2_EVENT_HMMU15_SECURITY_ERROR:
+		dcore = 2;
+		index_in_dcore = 1;
+	break;
+	default:
+		return ULONG_MAX;
+	}
+
+	return mmDCORE0_HMMU0_MMU_BASE + dcore * DCORE_OFFSET + index_in_dcore * DCORE_HMMU_OFFSET;
+}
+
 static int gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type, u64 *event_mask)
 {
 	bool is_pmmu = false;
 	u32 error_count = 0;
 	u64 mmu_base;
-	u8 index;

 	switch (event_type) {
-	case GAUDI2_EVENT_HMMU0_PAGE_FAULT_OR_WR_PERM ... GAUDI2_EVENT_HMMU3_SECURITY_ERROR:
-		index = (event_type - GAUDI2_EVENT_HMMU0_PAGE_FAULT_OR_WR_PERM) / 3;
-		mmu_base = mmDCORE0_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		break;
-	case GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_3_AXI_ERR_RSP:
-		index = (event_type - GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP);
-		mmu_base = mmDCORE0_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		break;
-	case GAUDI2_EVENT_HMMU8_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_HMMU11_SECURITY_ERROR:
-		index = (event_type - GAUDI2_EVENT_HMMU8_PAGE_FAULT_WR_PERM) / 3;
-		mmu_base = mmDCORE1_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		break;
-	case GAUDI2_EVENT_HMMU_8_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_11_AXI_ERR_RSP:
-		index = (event_type - GAUDI2_EVENT_HMMU_8_AXI_ERR_RSP);
-		mmu_base = mmDCORE1_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		break;
-	case GAUDI2_EVENT_HMMU7_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_HMMU4_SECURITY_ERROR:
-		index = (event_type - GAUDI2_EVENT_HMMU7_PAGE_FAULT_WR_PERM) / 3;
-		mmu_base = mmDCORE2_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		break;
-	case GAUDI2_EVENT_HMMU_7_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_4_AXI_ERR_RSP:
-		index = (event_type - GAUDI2_EVENT_HMMU_7_AXI_ERR_RSP);
-		mmu_base = mmDCORE2_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		break;
-	case GAUDI2_EVENT_HMMU15_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_HMMU12_SECURITY_ERROR:
-		index = (event_type - GAUDI2_EVENT_HMMU15_PAGE_FAULT_WR_PERM) / 3;
-		mmu_base = mmDCORE3_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		break;
-	case GAUDI2_EVENT_HMMU_15_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_12_AXI_ERR_RSP:
-		index = (event_type - GAUDI2_EVENT_HMMU_15_AXI_ERR_RSP);
-		mmu_base = mmDCORE3_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
+	case GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_12_AXI_ERR_RSP:
+	case GAUDI2_EVENT_HMMU0_SPI_BASE ... GAUDI2_EVENT_HMMU12_SECURITY_ERROR:
+		mmu_base = get_hmmu_base(event_type);
 		break;
+
 	case GAUDI2_EVENT_PMMU0_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_PMMU0_SECURITY_ERROR:
 	case GAUDI2_EVENT_PMMU_AXI_ERR_RSP_0:
 		is_pmmu = true;
@ -9025,6 +9172,9 @@ static int gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type,
 		return 0;
 	}

+	if (mmu_base == ULONG_MAX)
+		return 0;
+
 	error_count = gaudi2_handle_mmu_spi_sei_generic(hdev, event_type, mmu_base,
 							is_pmmu, event_mask);
 	hl_check_for_glbl_errors(hdev);
@ -9435,19 +9585,18 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;

 	case GAUDI2_EVENT_ARC_AXI_ERROR_RESPONSE_0:
-		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		error_count = gaudi2_handle_arc_farm_sei_err(hdev, event_type);
-		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
+		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;

 	case GAUDI2_EVENT_CPU_AXI_ERR_RSP:
 		error_count = gaudi2_handle_cpu_sei_err(hdev, event_type);
-		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
+		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
+		event_mask |= HL_NOTIFIER_EVENT_CRITICL_FW_ERR;
 		break;

 	case GAUDI2_EVENT_PDMA_CH0_AXI_ERR_RSP:
 	case GAUDI2_EVENT_PDMA_CH1_AXI_ERR_RSP:
-		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		error_count = gaudi2_handle_qm_sei_err(hdev, event_type, true, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
@ -9634,12 +9783,14 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent

 	case GAUDI2_EVENT_PCIE_DRAIN_COMPLETE:
 		error_count = gaudi2_handle_pcie_drain(hdev, &eq_entry->pcie_drain_ind_data);
+		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;

 	case GAUDI2_EVENT_PSOC59_RPM_ERROR_OR_DRAIN:
 		error_count = gaudi2_handle_psoc_drain(hdev,
 				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
+		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;

@ -9668,6 +9819,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 	case GAUDI2_EVENT_PSOC_AXI_ERR_RSP:
 		error_count = GAUDI2_NA_EVENT_CAUSE;
+		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_PSOC_PRSTN_FALL:
@ -9681,6 +9833,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 	case GAUDI2_EVENT_PCIE_FATAL_ERR:
 		error_count = GAUDI2_NA_EVENT_CAUSE;
+		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_TPC0_BMON_SPMU:
@ -9748,6 +9901,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_CPU_PKT_QUEUE_OUT_SYNC:
 		gaudi2_print_out_of_sync_info(hdev, event_type, &eq_entry->pkt_sync_err);
 		error_count = GAUDI2_NA_EVENT_CAUSE;
+		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;

@ -9789,6 +9943,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_CPU_PKT_SANITY_FAILED:
 		gaudi2_print_cpu_pkt_failure_info(hdev, event_type, &eq_entry->pkt_sync_err);
 		error_count = GAUDI2_NA_EVENT_CAUSE;
+		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;

--- a/drivers/accel/habanalabs/gaudi2/gaudi2P.h
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2P.h
@ -240,6 +240,8 @@
 #define GAUDI2_SOB_INCREMENT_BY_ONE	(FIELD_PREP(DCORE0_SYNC_MNGR_OBJS_SOB_OBJ_VAL_MASK, 1) | \
 					FIELD_PREP(DCORE0_SYNC_MNGR_OBJS_SOB_OBJ_INC_MASK, 1))

+#define GAUDI2_NUM_TESTED_QS (GAUDI2_QUEUE_ID_CPU_PQ - GAUDI2_QUEUE_ID_PDMA_0_0)
+
 #define GAUDI2_NUM_OF_GLBL_ERR_CAUSE		8

 enum gaudi2_reserved_sob_id {
@ -452,6 +454,17 @@ struct dup_block_ctx {
 	unsigned int instances;
 };

+/**
+ * struct gaudi2_queues_test_info - Holds the address of a the messages used for testing the
+ *                                  device queues.
+ * @dma_addr: the address used by the HW for accessing the message.
+ * @kern_addr: The address used by the driver for accessing the message.
+ */
+struct gaudi2_queues_test_info {
+	dma_addr_t dma_addr;
+	void *kern_addr;
+};
+
 /**
 * struct gaudi2_device - ASIC specific manage structure.
 * @cpucp_info_get: get information on device from CPU-CP
@ -510,6 +523,7 @@ struct dup_block_ctx {
 * @flush_db_fifo: flag to force flush DB FIFO after a write.
 * @hbm_cfg: HBM subsystem settings
 * @hw_queues_lock_mutex: used by simulator instead of hw_queues_lock.
+ * @queues_test_info: information used by the driver when testing the HW queues.
 */
 struct gaudi2_device {
 	int (*cpucp_info_get)(struct hl_device *hdev);
@ -537,6 +551,9 @@ struct gaudi2_device {
 	u32				events_stat[GAUDI2_EVENT_SIZE];
 	u32				events_stat_aggregate[GAUDI2_EVENT_SIZE];
 	u32				num_of_valid_hw_events;
+
+	/* Queue testing */
+	struct gaudi2_queues_test_info	queues_test_info[GAUDI2_NUM_TESTED_QS];
 };

 /*
--- a/drivers/accel/habanalabs/goya/goya.c
+++ b/drivers/accel/habanalabs/goya/goya.c
@ -473,6 +473,7 @@ int goya_set_fixed_properties(struct hl_device *hdev)

 	prop->first_available_user_interrupt = USHRT_MAX;
 	prop->tpc_interrupt_id = USHRT_MAX;
+	prop->eq_interrupt_id = GOYA_EVENT_QUEUE_MSIX_IDX;

 	for (i = 0 ; i < HL_MAX_DCORES ; i++)
 		prop->first_available_cq[i] = USHRT_MAX;
--- a/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h
+++ b/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h
@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0
 *
- * Copyright 2020-2022 HabanaLabs, Ltd.
+ * Copyright 2020-2023 HabanaLabs, Ltd.
 * All Rights Reserved.
 *
 */
@ -543,6 +543,8 @@
 #define HBM_MC_SPI_IEEE1500_COMP_MASK		BIT(3)
 #define HBM_MC_SPI_IEEE1500_PAUSED_MASK		BIT(4)

+#define ARC_FARM_OFFSET (mmARC_FARM_ARC1_AUX_BASE - mmARC_FARM_ARC0_AUX_BASE)
+
 #include "nic0_qpc0_regs.h"
 #include "nic0_qm0_regs.h"
 #include "nic0_qm_arc_aux0_regs.h"
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@ -77,7 +77,8 @@ amdgpu-y += \
 	vi.o mxgpu_vi.o nbio_v6_1.o soc15.o emu_soc.o mxgpu_ai.o nbio_v7_0.o vega10_reg_init.o \
 	vega20_reg_init.o nbio_v7_4.o nbio_v2_3.o nv.o arct_reg_init.o mxgpu_nv.o \
 	nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o \
-	sienna_cichlid.o smu_v13_0_10.o nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o
+	sienna_cichlid.o smu_v13_0_10.o nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o \
+	nbio_v7_9.o

 # add DF block
 amdgpu-y += \
@ -92,7 +93,7 @@ amdgpu-y += \
 	gfxhub_v1_0.o mmhub_v1_0.o gmc_v9_0.o gfxhub_v1_1.o mmhub_v9_4.o \
 	gfxhub_v2_0.o mmhub_v2_0.o gmc_v10_0.o gfxhub_v2_1.o mmhub_v2_3.o \
 	mmhub_v1_7.o gfxhub_v3_0.o mmhub_v3_0.o mmhub_v3_0_2.o gmc_v11_0.o \
-	mmhub_v3_0_1.o gfxhub_v3_0_3.o
+	mmhub_v3_0_1.o gfxhub_v3_0_3.o gfxhub_v1_2.o mmhub_v1_8.o

 # add UMC block
 amdgpu-y += \
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@ -3181,9 +3181,11 @@ static int amdgpu_device_ip_reinit_late_sriov(struct amdgpu_device *adev)
 		AMD_IP_BLOCK_TYPE_DCE,
 		AMD_IP_BLOCK_TYPE_GFX,
 		AMD_IP_BLOCK_TYPE_SDMA,
+		AMD_IP_BLOCK_TYPE_MES,
 		AMD_IP_BLOCK_TYPE_UVD,
 		AMD_IP_BLOCK_TYPE_VCE,
-		AMD_IP_BLOCK_TYPE_VCN
+		AMD_IP_BLOCK_TYPE_VCN,
+		AMD_IP_BLOCK_TYPE_JPEG
 	};

 	for (i = 0; i < ARRAY_SIZE(ip_order); i++) {
@ -5176,6 +5178,7 @@ static inline void amdgpu_device_stop_pending_resets(struct amdgpu_device *adev)
 *
 * @adev: amdgpu_device pointer
 * @job: which job trigger hang
+ * @reset_context: amdgpu reset context pointer
 *
 * Attempt to reset the GPU if it has hung (all asics).
 * Attempt to do soft-reset or full-reset and reinitialize Asic
@ -5345,8 +5348,9 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 		if (r)
 			adev->asic_reset_res = r;

-		/* Aldebaran supports ras in SRIOV, so need resume ras during reset */
-		if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2))
+		/* Aldebaran and gfx_11_0_3 support ras in SRIOV, so need resume ras during reset */
+		if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ||
+		    adev->ip_versions[GC_HWIP][0] == IP_VERSION(11, 0, 3))
 			amdgpu_ras_resume(adev);
 	} else {
 		r = amdgpu_do_asic_reset(device_list_handle, reset_context);
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@ -37,6 +37,7 @@
 #include "nbio_v6_1.h"
 #include "nbio_v7_0.h"
 #include "nbio_v7_4.h"
+#include "nbio_v7_9.h"
 #include "hdp_v4_0.h"
 #include "vega10_ih.h"
 #include "vega20_ih.h"
@ -1546,6 +1547,7 @@ static int amdgpu_discovery_set_gmc_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(9, 4, 0):
 	case IP_VERSION(9, 4, 1):
 	case IP_VERSION(9, 4, 2):
+	case IP_VERSION(9, 4, 3):
 		amdgpu_device_ip_block_add(adev, &gmc_v9_0_ip_block);
 		break;
 	case IP_VERSION(10, 1, 10):
@ -1948,9 +1950,8 @@ static int amdgpu_discovery_set_mm_ip_blocks(struct amdgpu_device *adev)
 		case IP_VERSION(4, 0, 2):
 		case IP_VERSION(4, 0, 4):
 			amdgpu_device_ip_block_add(adev, &vcn_v4_0_ip_block);
-			if (!amdgpu_sriov_vf(adev))
-				amdgpu_device_ip_block_add(adev, &jpeg_v4_0_ip_block);
-			break;
+			amdgpu_device_ip_block_add(adev, &jpeg_v4_0_ip_block);
+			return 0;
 		default:
 			dev_err(adev->dev,
 				"Failed to add vcn/jpeg ip block(UVD_HWIP:0x%x)\n",
@ -2181,6 +2182,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(9, 4, 0):
 	case IP_VERSION(9, 4, 1):
 	case IP_VERSION(9, 4, 2):
+	case IP_VERSION(9, 4, 3):
 		adev->family = AMDGPU_FAMILY_AI;
 		break;
 	case IP_VERSION(9, 1, 0):
@ -2265,6 +2267,10 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		adev->nbio.funcs = &nbio_v7_4_funcs;
 		adev->nbio.hdp_flush_reg = &nbio_v7_4_hdp_flush_reg;
 		break;
+	case IP_VERSION(7, 9, 0):
+		adev->nbio.funcs = &nbio_v7_9_funcs;
+		adev->nbio.hdp_flush_reg = &nbio_v7_9_hdp_flush_reg;
+		break;
 	case IP_VERSION(7, 2, 0):
 	case IP_VERSION(7, 2, 1):
 	case IP_VERSION(7, 3, 0):
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@ -148,8 +148,8 @@ uint amdgpu_pcie_lane_cap;
 u64 amdgpu_cg_mask = 0xffffffffffffffff;
 uint amdgpu_pg_mask = 0xffffffff;
 uint amdgpu_sdma_phase_quantum = 32;
-char *amdgpu_disable_cu = NULL;
-char *amdgpu_virtual_display = NULL;
+char *amdgpu_disable_cu;
+char *amdgpu_virtual_display;

 /*
 * OverDrive(bit 14) disabled by default
@ -822,7 +822,7 @@ MODULE_PARM_DESC(no_system_mem_limit, "disable system memory limit (false = defa
 * DOC: no_queue_eviction_on_vm_fault (int)
 * If set, process queues will not be evicted on gpuvm fault. This is to keep the wavefront context for debugging (0 = queue eviction, 1 = no queue eviction). The default is 0 (queue eviction).
 */
-int amdgpu_no_queue_eviction_on_vm_fault = 0;
+int amdgpu_no_queue_eviction_on_vm_fault;
 MODULE_PARM_DESC(no_queue_eviction_on_vm_fault, "No queue eviction on VM fault (0 = queue eviction, 1 = no queue eviction)");
 module_param_named(no_queue_eviction_on_vm_fault, amdgpu_no_queue_eviction_on_vm_fault, int, 0444);
 #endif
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@ -624,6 +624,7 @@ void amdgpu_gmc_noretry_set(struct amdgpu_device *adev)
 				gc_ver == IP_VERSION(9, 4, 0) ||
 				gc_ver == IP_VERSION(9, 4, 1) ||
 				gc_ver == IP_VERSION(9, 4, 2) ||
+				gc_ver == IP_VERSION(9, 4, 3) ||
 				gc_ver >= IP_VERSION(10, 3, 0));

 	gmc->noretry = (amdgpu_noretry == -1) ? noretry_default : amdgpu_noretry;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@ -118,6 +118,10 @@ int amdgpu_jpeg_dec_ring_test_ring(struct amdgpu_ring *ring)
 	unsigned i;
 	int r;

+	/* JPEG in SRIOV does not support direct register read/write */
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	WREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch, 0xCAFEDEAD);
 	r = amdgpu_ring_alloc(ring, 3);
 	if (r)
@ -202,17 +206,18 @@ int amdgpu_jpeg_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	} else {
 		r = 0;
 	}
+	if (!amdgpu_sriov_vf(adev)) {
+		for (i = 0; i < adev->usec_timeout; i++) {
+			tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch);
+			if (tmp == 0xDEADBEEF)
+				break;
+			udelay(1);
+		}

-	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch);
-		if (tmp == 0xDEADBEEF)
-			break;
-		udelay(1);
+		if (i >= adev->usec_timeout)
+			r = -ETIMEDOUT;
 	}

-	if (i >= adev->usec_timeout)
-		r = -ETIMEDOUT;
-
 	dma_fence_put(fence);
 error:
 	return r;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@ -1104,6 +1104,11 @@ int amdgpu_mes_ctx_alloc_meta_data(struct amdgpu_device *adev,
 			    &ctx_data->meta_data_obj,
 			    &ctx_data->meta_data_mc_addr,
 			    &ctx_data->meta_data_ptr);
+	if (r) {
+		dev_warn(adev->dev, "(%d) create CTX bo failed\n", r);
+		return r;
+	}
+
 	if (!ctx_data->meta_data_obj)
 		return -ENOMEM;

@ -1328,12 +1333,9 @@ int amdgpu_mes_self_test(struct amdgpu_device *adev)
 	struct amdgpu_mes_ctx_data ctx_data = {0};
 	struct amdgpu_ring *added_rings[AMDGPU_MES_CTX_MAX_RINGS] = { NULL };
 	int gang_ids[3] = {0};
-	int queue_types[][2] = { { AMDGPU_RING_TYPE_GFX,
-				   AMDGPU_MES_CTX_MAX_GFX_RINGS},
-				 { AMDGPU_RING_TYPE_COMPUTE,
-				   AMDGPU_MES_CTX_MAX_COMPUTE_RINGS},
-				 { AMDGPU_RING_TYPE_SDMA,
-				   AMDGPU_MES_CTX_MAX_SDMA_RINGS } };
+	int queue_types[][2] = { { AMDGPU_RING_TYPE_GFX, 1 },
+				 { AMDGPU_RING_TYPE_COMPUTE, 1 },
+				 { AMDGPU_RING_TYPE_SDMA, 1} };
 	int i, r, pasid, k = 0;

 	pasid = amdgpu_pasid_alloc(16);
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@ -148,6 +148,7 @@ static int psp_init_sriov_microcode(struct psp_context *psp)
 		break;
 	case IP_VERSION(13, 0, 10):
 		adev->virt.autoload_ucode_id = AMDGPU_UCODE_ID_CP_MES1_DATA;
+		ret = psp_init_cap_microcode(psp, ucode_prefix);
 		break;
 	default:
 		return -EINVAL;
@ -838,7 +839,15 @@ static void psp_prep_tmr_unload_cmd_buf(struct psp_context *psp,
 static int psp_tmr_unload(struct psp_context *psp)
 {
 	int ret;
-	struct psp_gfx_cmd_resp *cmd = acquire_psp_cmd_buf(psp);
+	struct psp_gfx_cmd_resp *cmd;
+
+	/* skip TMR unload for Navi12 and CHIP_SIENNA_CICHLID SRIOV,
+	 * as TMR is not loaded at all
+	 */
+	if (amdgpu_sriov_vf(psp->adev) && psp_skip_tmr(psp))
+		return 0;
+
+	cmd = acquire_psp_cmd_buf(psp);

 	psp_prep_tmr_unload_cmd_buf(psp, cmd);
 	dev_dbg(psp->adev->dev, "free PSP TMR buffer\n");
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@ -34,6 +34,7 @@
 #include "amdgpu_atomfirmware.h"
 #include "amdgpu_xgmi.h"
 #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"
+#include "nbio_v4_3.h"
 #include "atom.h"
 #include "amdgpu_reset.h"

@ -2340,6 +2341,7 @@ static bool amdgpu_ras_asic_supported(struct amdgpu_device *adev)
 	if (amdgpu_sriov_vf(adev)) {
 		switch (adev->ip_versions[MP0_HWIP][0]) {
 		case IP_VERSION(13, 0, 2):
+		case IP_VERSION(13, 0, 10):
 			return true;
 		default:
 			return false;
@ -2561,6 +2563,16 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 		if (!adev->gmc.xgmi.connected_to_cpu)
 			adev->nbio.ras = &nbio_v7_4_ras;
 		break;
+	case IP_VERSION(4, 3, 0):
+		if (adev->ras_hw_enabled & (1 << AMDGPU_RAS_BLOCK__DF))
+			/* unlike other generation of nbio ras,
+			 * nbio v4_3 only support fatal error interrupt
+			 * to inform software that DF is freezed due to
+			 * system fatal error event. driver should not
+			 * enable nbio ras in such case. Instead,
+			 * check DF RAS */
+			adev->nbio.ras = &nbio_v4_3_ras;
+		break;
 	default:
 		/* nbio ras is not available */
 		break;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@ -181,14 +181,14 @@ static bool __get_eeprom_i2c_addr(struct amdgpu_device *adev,
 	switch (adev->asic_type) {
 	case CHIP_VEGA20:
 		control->i2c_address = EEPROM_I2C_MADDR_0;
-		break;
+		return true;

 	case CHIP_ARCTURUS:
 		return __get_eeprom_i2c_addr_arct(adev, control);

 	case CHIP_SIENNA_CICHLID:
 		control->i2c_address = EEPROM_I2C_MADDR_0;
-		break;
+		return true;

 	case CHIP_ALDEBARAN:
 		if (strnstr(atom_ctx->vbios_version, "D673",
@ -196,7 +196,7 @@ static bool __get_eeprom_i2c_addr(struct amdgpu_device *adev,
 			control->i2c_address = EEPROM_I2C_MADDR_4;
 		else
 			control->i2c_address = EEPROM_I2C_MADDR_0;
-		break;
+		return true;

 	case CHIP_IP_DISCOVERY:
 		return __get_eeprom_i2c_addr_ip_discovery(adev, control);
@ -204,17 +204,6 @@ static bool __get_eeprom_i2c_addr(struct amdgpu_device *adev,
 	default:
 		return false;
 	}
-
-	switch (adev->ip_versions[MP1_HWIP][0]) {
-	case IP_VERSION(13, 0, 0):
-		control->i2c_address = EEPROM_I2C_MADDR_4;
-		break;
-
-	default:
-		break;
-	}
-
-	return true;
 }

 static void
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
@ -126,19 +126,6 @@ void amdgpu_ucode_print_gfx_hdr(const struct common_firmware_header *hdr)
 	}
 }

-void amdgpu_ucode_print_imu_hdr(const struct common_firmware_header *hdr)
-{
-	uint16_t version_major = le16_to_cpu(hdr->header_version_major);
-	uint16_t version_minor = le16_to_cpu(hdr->header_version_minor);
-
-	DRM_DEBUG("IMU\n");
-	amdgpu_ucode_print_common_hdr(hdr);
-
-	if (version_major != 1) {
-		DRM_ERROR("Unknown GFX ucode version: %u.%u\n", version_major, version_minor);
-	}
-}
-
 void amdgpu_ucode_print_rlc_hdr(const struct common_firmware_header *hdr)
 {
 	uint16_t version_major = le16_to_cpu(hdr->header_version_major);
@ -472,6 +459,12 @@ void amdgpu_ucode_print_psp_hdr(const struct common_firmware_header *hdr)
 				DRM_DEBUG("psp_dbg_drv_size_bytes: %u\n",
 					  le32_to_cpu(desc->size_bytes));
 				break;
+			case PSP_FW_TYPE_PSP_RAS_DRV:
+				DRM_DEBUG("psp_ras_drv_version: %u\n",
+					  le32_to_cpu(desc->fw_version));
+				DRM_DEBUG("psp_ras_drv_size_bytes: %u\n",
+					  le32_to_cpu(desc->size_bytes));
+				break;
 			default:
 				DRM_DEBUG("Unsupported PSP fw type: %d\n", desc->fw_type);
 				break;
@ -669,6 +662,8 @@ const char *amdgpu_ucode_name(enum AMDGPU_UCODE_ID ucode_id)
 		return "VCN1_RAM";
 	case AMDGPU_UCODE_ID_DMCUB:
 		return "DMCUB";
+	case AMDGPU_UCODE_ID_CAP:
+		return "CAP";
 	default:
 		return "UNKNOWN UCODE";
 	}
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
@ -232,7 +232,7 @@ int amdgpu_umc_ras_sw_init(struct amdgpu_device *adev)
 	if (!ras->ras_block.ras_late_init)
 		ras->ras_block.ras_late_init = amdgpu_umc_ras_late_init;

-	if (ras->ras_block.ras_cb)
+	if (!ras->ras_block.ras_cb)
 		ras->ras_block.ras_cb = amdgpu_umc_process_ras_data_cb;

 	return 0;
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@ -585,6 +585,7 @@ err:
 /**
 * amdgpu_vce_validate_bo - make sure not to cross 4GB boundary
 *
+ * @p: cs parser
 * @ib: indirect buffer to use
 * @lo: address of lower dword
 * @hi: address of higher dword
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@ -673,6 +673,7 @@ void amdgpu_vm_pt_free_work(struct work_struct *work)
 * @adev: amdgpu device structure
 * @vm: amdgpu vm structure
 * @start: optional cursor where to start freeing PDs/PTs
+ * @unlocked: vm resv unlock status
 *
 * Free the page directory or page table level and all sub levels.
 */
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
@ -171,7 +171,7 @@ static void amdgpu_vm_sdma_copy_ptes(struct amdgpu_vm_update_params *p,

 	src += p->num_dw_left * 4;

-	pe += amdgpu_gmc_sign_extend(amdgpu_bo_gpu_offset_no_check(bo));
+	pe += amdgpu_bo_gpu_offset_no_check(bo);
 	trace_amdgpu_vm_copy_ptes(pe, src, count, p->immediate);

 	amdgpu_vm_copy_pte(p->adev, ib, pe, src, count);
@ -198,7 +198,7 @@ static void amdgpu_vm_sdma_set_ptes(struct amdgpu_vm_update_params *p,
 {
 	struct amdgpu_ib *ib = p->job->ibs;

-	pe += amdgpu_gmc_sign_extend(amdgpu_bo_gpu_offset_no_check(bo));
+	pe += amdgpu_bo_gpu_offset_no_check(bo);
 	trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags, p->immediate);
 	if (count < 3) {
 		amdgpu_vm_write_pte(p->adev, ib, pe, addr | flags,
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@ -1068,7 +1068,7 @@ int amdgpu_xgmi_ras_sw_init(struct amdgpu_device *adev)
 		return err;
 	}

-	strcpy(ras->ras_block.ras_comm.name, "xgmi_wafl_pcs");
+	strcpy(ras->ras_block.ras_comm.name, "xgmi_wafl");
 	ras->ras_block.ras_comm.block = AMDGPU_RAS_BLOCK__XGMI_WAFL;
 	ras->ras_block.ras_comm.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE;
 	adev->gmc.xgmi.ras_if = &ras->ras_block.ras_comm;
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@ -0,0 +1,471 @@
+/*
+ * Copyright 2022 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 "amdgpu.h"
+#include "gfxhub_v1_2.h"
+#include "gfxhub_v1_1.h"
+
+#include "gc/gc_9_4_3_offset.h"
+#include "gc/gc_9_4_3_sh_mask.h"
+#include "vega10_enum.h"
+
+#include "soc15_common.h"
+
+#define regVM_L2_CNTL3_DEFAULT	0x80100007
+#define regVM_L2_CNTL4_DEFAULT	0x000000c1
+
+static u64 gfxhub_v1_2_get_mc_fb_offset(struct amdgpu_device *adev)
+{
+	return (u64)RREG32_SOC15(GC, 0, regMC_VM_FB_OFFSET) << 24;
+}
+
+static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
+					 uint32_t vmid,
+					 uint64_t page_table_base)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+
+	WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+			    hub->ctx_addr_distance * vmid,
+			    lower_32_bits(page_table_base));
+
+	WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+			    hub->ctx_addr_distance * vmid,
+			    upper_32_bits(page_table_base));
+}
+
+static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
+{
+	uint64_t pt_base;
+
+	if (adev->gmc.pdb0_bo)
+		pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
+	else
+		pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+	gfxhub_v1_2_setup_vm_pt_regs(adev, 0, pt_base);
+
+	/* If use GART for FB translation, vmid0 page table covers both
+	 * vram and system memory (gart)
+	 */
+	if (adev->gmc.pdb0_bo) {
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+				(u32)(adev->gmc.fb_start >> 12));
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+				(u32)(adev->gmc.fb_start >> 44));
+
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+				(u32)(adev->gmc.gart_end >> 12));
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+				(u32)(adev->gmc.gart_end >> 44));
+	} else {
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+				(u32)(adev->gmc.gart_start >> 12));
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+				(u32)(adev->gmc.gart_start >> 44));
+
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+				(u32)(adev->gmc.gart_end >> 12));
+		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+				(u32)(adev->gmc.gart_end >> 44));
+	}
+}
+
+static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
+{
+	uint64_t value;
+	uint32_t tmp;
+
+	/* Program the AGP BAR */
+	WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_BASE, 0);
+	WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
+	WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24);
+
+	if (!amdgpu_sriov_vf(adev) || adev->asic_type <= CHIP_VEGA10) {
+		/* Program the system aperture low logical page number. */
+		WREG32_SOC15_RLC(GC, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
+			min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
+
+		if (adev->apu_flags & AMD_APU_IS_RAVEN2)
+			/*
+			* Raven2 has a HW issue that it is unable to use the
+			* vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR.
+			* So here is the workaround that increase system
+			* aperture high address (add 1) to get rid of the VM
+			* fault and hardware hang.
+			*/
+			WREG32_SOC15_RLC(GC, 0,
+					 regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
+					 max((adev->gmc.fb_end >> 18) + 0x1,
+					     adev->gmc.agp_end >> 18));
+		else
+			WREG32_SOC15_RLC(GC, 0,
+				regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
+				max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
+
+		/* Set default page address. */
+		value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
+		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
+			     (u32)(value >> 12));
+		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
+			     (u32)(value >> 44));
+
+		/* Program "protection fault". */
+		WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
+			     (u32)(adev->dummy_page_addr >> 12));
+		WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
+			     (u32)((u64)adev->dummy_page_addr >> 44));
+
+		tmp = RREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL2);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2,
+				    ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
+		WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+	}
+
+	/* In the case squeezing vram into GART aperture, we don't use
+	 * FB aperture and AGP aperture. Disable them.
+	 */
+	if (adev->gmc.pdb0_bo) {
+		WREG32_SOC15(GC, 0, regMC_VM_FB_LOCATION_TOP, 0);
+		WREG32_SOC15(GC, 0, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF);
+		WREG32_SOC15(GC, 0, regMC_VM_AGP_TOP, 0);
+		WREG32_SOC15(GC, 0, regMC_VM_AGP_BOT, 0xFFFFFF);
+		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF);
+		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
+	}
+}
+
+static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+
+	/* Setup TLB control */
+	tmp = RREG32_SOC15(GC, 0, regMC_VM_MX_L1_TLB_CNTL);
+
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+			    ENABLE_ADVANCED_DRIVER_MODEL, 1);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+			    SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+			    MTYPE, MTYPE_UC);/* XXX for emulation. */
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
+
+	WREG32_SOC15_RLC(GC, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
+}
+
+static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+
+	/* Setup L2 cache */
+	tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1);
+	/* XXX for emulation, Refer to closed source code.*/
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE,
+			    0);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
+	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL, tmp);
+
+	tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL2);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
+	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL2, tmp);
+
+	tmp = regVM_L2_CNTL3_DEFAULT;
+	if (adev->gmc.translate_further) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+	} else {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+	}
+	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL3, tmp);
+
+	tmp = regVM_L2_CNTL4_DEFAULT;
+	if (adev->gmc.xgmi.connected_to_cpu) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
+	} else {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
+	}
+	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL4, tmp);
+}
+
+static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+
+	tmp = RREG32_SOC15(GC, 0, regVM_CONTEXT0_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
+			adev->gmc.vmid0_page_table_depth);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE,
+			adev->gmc.vmid0_page_table_block_size);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL,
+			    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+	WREG32_SOC15(GC, 0, regVM_CONTEXT0_CNTL, tmp);
+}
+
+static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
+{
+	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
+		     0XFFFFFFFF);
+	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32,
+		     0x0000000F);
+
+	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32,
+		     0);
+	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32,
+		     0);
+
+	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0);
+	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0);
+
+}
+
+static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	unsigned num_level, block_size;
+	uint32_t tmp;
+	int i;
+
+	num_level = adev->vm_manager.num_level;
+	block_size = adev->vm_manager.block_size;
+	if (adev->gmc.translate_further)
+		num_level -= 1;
+	else
+		block_size -= 9;
+
+	for (i = 0; i <= 14; i++) {
+		tmp = RREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT1_CNTL, i);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
+				    num_level);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
+				    1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    PAGE_TABLE_BLOCK_SIZE,
+				    block_size);
+		/* Send no-retry XNACK on fault to suppress VM fault storm.
+		 * On Aldebaran, XNACK can be enabled in the SQ per-process.
+		 * Retry faults need to be enabled for that to work.
+		 */
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
+				    !adev->gmc.noretry ||
+				    adev->asic_type == CHIP_ALDEBARAN);
+		WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT1_CNTL,
+				    i * hub->ctx_distance, tmp);
+		WREG32_SOC15_OFFSET(GC, 0,
+				    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
+				    i * hub->ctx_addr_distance, 0);
+		WREG32_SOC15_OFFSET(GC, 0,
+				    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
+				    i * hub->ctx_addr_distance, 0);
+		WREG32_SOC15_OFFSET(GC, 0,
+				    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
+				    i * hub->ctx_addr_distance,
+				    lower_32_bits(adev->vm_manager.max_pfn - 1));
+		WREG32_SOC15_OFFSET(GC, 0,
+				    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
+				    i * hub->ctx_addr_distance,
+				    upper_32_bits(adev->vm_manager.max_pfn - 1));
+	}
+}
+
+static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	unsigned i;
+
+	for (i = 0 ; i < 18; ++i) {
+		WREG32_SOC15_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
+				    i * hub->eng_addr_distance, 0xffffffff);
+		WREG32_SOC15_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
+				    i * hub->eng_addr_distance, 0x1f);
+	}
+}
+
+static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
+{
+	if (amdgpu_sriov_vf(adev) && adev->asic_type != CHIP_ARCTURUS) {
+		/*
+		 * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are
+		 * VF copy registers so vbios post doesn't program them, for
+		 * SRIOV driver need to program them
+		 */
+		WREG32_SOC15_RLC(GC, 0, regMC_VM_FB_LOCATION_BASE,
+			     adev->gmc.vram_start >> 24);
+		WREG32_SOC15_RLC(GC, 0, regMC_VM_FB_LOCATION_TOP,
+			     adev->gmc.vram_end >> 24);
+	}
+
+	/* GART Enable. */
+	gfxhub_v1_2_init_gart_aperture_regs(adev);
+	gfxhub_v1_2_init_system_aperture_regs(adev);
+	gfxhub_v1_2_init_tlb_regs(adev);
+	if (!amdgpu_sriov_vf(adev))
+		gfxhub_v1_2_init_cache_regs(adev);
+
+	gfxhub_v1_2_enable_system_domain(adev);
+	if (!amdgpu_sriov_vf(adev))
+		gfxhub_v1_2_disable_identity_aperture(adev);
+	gfxhub_v1_2_setup_vmid_config(adev);
+	gfxhub_v1_2_program_invalidation(adev);
+
+	return 0;
+}
+
+static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	u32 tmp;
+	u32 i;
+
+	/* Disable all tables */
+	for (i = 0; i < 16; i++)
+		WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_CNTL,
+				    i * hub->ctx_distance, 0);
+
+	/* Setup TLB control */
+	tmp = RREG32_SOC15(GC, 0, regMC_VM_MX_L1_TLB_CNTL);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0);
+	tmp = REG_SET_FIELD(tmp,
+				MC_VM_MX_L1_TLB_CNTL,
+				ENABLE_ADVANCED_DRIVER_MODEL,
+				0);
+	WREG32_SOC15_RLC(GC, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
+
+	/* Setup L2 cache */
+	tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0);
+	WREG32_SOC15(GC, 0, regVM_L2_CNTL, tmp);
+	WREG32_SOC15(GC, 0, regVM_L2_CNTL3, 0);
+}
+
+/**
+ * gfxhub_v1_2_set_fault_enable_default - update GART/VM fault handling
+ *
+ * @adev: amdgpu_device pointer
+ * @value: true redirects VM faults to the default page
+ */
+static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
+						 bool value)
+{
+	u32 tmp;
+	tmp = RREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp,
+			VM_L2_PROTECTION_FAULT_CNTL,
+			TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
+			value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	if (!value) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				CRASH_ON_NO_RETRY_FAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				CRASH_ON_RETRY_FAULT, 1);
+	}
+	WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL, tmp);
+}
+
+static void gfxhub_v1_2_init(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+
+	hub->ctx0_ptb_addr_lo32 =
+		SOC15_REG_OFFSET(GC, 0,
+				regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
+	hub->ctx0_ptb_addr_hi32 =
+		SOC15_REG_OFFSET(GC, 0,
+				regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
+	hub->vm_inv_eng0_sem =
+		SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_SEM);
+	hub->vm_inv_eng0_req =
+		SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_REQ);
+	hub->vm_inv_eng0_ack =
+		SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ACK);
+	hub->vm_context0_cntl =
+		SOC15_REG_OFFSET(GC, 0, regVM_CONTEXT0_CNTL);
+	hub->vm_l2_pro_fault_status =
+		SOC15_REG_OFFSET(GC, 0, regVM_L2_PROTECTION_FAULT_STATUS);
+	hub->vm_l2_pro_fault_cntl =
+		SOC15_REG_OFFSET(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL);
+
+	hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL;
+	hub->ctx_addr_distance = regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
+		regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
+	hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - regVM_INVALIDATE_ENG0_REQ;
+	hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
+		regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+}
+
+
+const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs = {
+	.get_mc_fb_offset = gfxhub_v1_2_get_mc_fb_offset,
+	.setup_vm_pt_regs = gfxhub_v1_2_setup_vm_pt_regs,
+	.gart_enable = gfxhub_v1_2_gart_enable,
+	.gart_disable = gfxhub_v1_2_gart_disable,
+	.set_fault_enable_default = gfxhub_v1_2_set_fault_enable_default,
+	.init = gfxhub_v1_2_init,
+	.get_xgmi_info = gfxhub_v1_1_get_xgmi_info,
+};
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h
@ -0,0 +1,29 @@
+/*
+ * Copyright 2022 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.
+ *
+ */
+
+#ifndef __GFXHUB_V1_2_H__
+#define __GFXHUB_V1_2_H__
+
+extern const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs;
+
+#endif
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@ -274,6 +274,8 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 *
 * @adev: amdgpu_device pointer
 * @vmid: vm instance to flush
+ * @vmhub: which hub to flush
+ * @flush_type: the flush type
 *
 * Flush the TLB for the requested page table.
 */
@ -313,6 +315,8 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 *
 * @adev: amdgpu_device pointer
 * @pasid: pasid to be flush
+ * @flush_type: the flush type
+ * @all_hub: flush all hubs
 *
 * Flush the TLB for the requested pasid.
 */
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@ -49,8 +49,10 @@
 #include "mmhub_v1_0.h"
 #include "athub_v1_0.h"
 #include "gfxhub_v1_1.h"
+#include "gfxhub_v1_2.h"
 #include "mmhub_v9_4.h"
 #include "mmhub_v1_7.h"
+#include "mmhub_v1_8.h"
 #include "umc_v6_1.h"
 #include "umc_v6_0.h"
 #include "umc_v6_7.h"
@ -657,6 +659,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 		case IP_VERSION(2, 4, 0):
 			mmhub_cid = mmhub_client_ids_renoir[cid][rw];
 			break;
+		case IP_VERSION(1, 8, 0):
 		case IP_VERSION(9, 4, 2):
 			mmhub_cid = mmhub_client_ids_aldebaran[cid][rw];
 			break;
@ -735,7 +738,8 @@ static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid,
 static bool gmc_v9_0_use_invalidate_semaphore(struct amdgpu_device *adev,
 				       uint32_t vmhub)
 {
-	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2))
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ||
+	    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
 		return false;

 	return ((vmhub == AMDGPU_MMHUB_0 ||
@ -1144,6 +1148,7 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 1):
 	case IP_VERSION(9, 4, 2):
+	case IP_VERSION(9, 4, 3):
 		if (is_vram) {
 			if (bo_adev == adev) {
 				if (uncached)
@ -1155,8 +1160,8 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 				/* FIXME: is this still needed? Or does
 				 * amdgpu_ttm_tt_pde_flags already handle this?
 				 */
-				if (adev->ip_versions[GC_HWIP][0] ==
-					IP_VERSION(9, 4, 2) &&
+				if ((adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ||
+				     adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) &&
 				    adev->gmc.xgmi.connected_to_cpu)
 					snoop = true;
 			} else {
@ -1329,6 +1334,9 @@ static void gmc_v9_0_set_mmhub_funcs(struct amdgpu_device *adev)
 	case IP_VERSION(9, 4, 2):
 		adev->mmhub.funcs = &mmhub_v1_7_funcs;
 		break;
+	case IP_VERSION(1, 8, 0):
+		adev->mmhub.funcs = &mmhub_v1_8_funcs;
+		break;
 	default:
 		adev->mmhub.funcs = &mmhub_v1_0_funcs;
 		break;
@ -1355,7 +1363,10 @@ static void gmc_v9_0_set_mmhub_ras_funcs(struct amdgpu_device *adev)

 static void gmc_v9_0_set_gfxhub_funcs(struct amdgpu_device *adev)
 {
-	adev->gfxhub.funcs = &gfxhub_v1_0_funcs;
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
+		adev->gfxhub.funcs = &gfxhub_v1_2_funcs;
+	else
+		adev->gfxhub.funcs = &gfxhub_v1_0_funcs;
 }

 static void gmc_v9_0_set_hdp_ras_funcs(struct amdgpu_device *adev)
@ -1544,6 +1555,7 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev)
 		case IP_VERSION(9, 4, 0):
 		case IP_VERSION(9, 4, 1):
 		case IP_VERSION(9, 4, 2):
+		case IP_VERSION(9, 4, 3):
 		default:
 			adev->gmc.gart_size = 512ULL << 20;
 			break;
@ -1673,6 +1685,7 @@ static int gmc_v9_0_sw_init(void *handle)
 	case IP_VERSION(9, 4, 0):
 	case IP_VERSION(9, 3, 0):
 	case IP_VERSION(9, 4, 2):
+	case IP_VERSION(9, 4, 3):
 		adev->num_vmhubs = 2;


@ -1769,7 +1782,8 @@ static int gmc_v9_0_sw_init(void *handle)
 	 */
 	adev->vm_manager.first_kfd_vmid =
 		(adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 1) ||
-		 adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2)) ? 3 : 8;
+		 adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ||
+		 adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) ? 3 : 8;

 	amdgpu_vm_manager_init(adev);

--- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
@ -119,7 +119,7 @@ force_update_wptr_for_self_int(struct amdgpu_device *adev,
 * ih_v6_0_toggle_ring_interrupts - toggle the interrupt ring buffer
 *
 * @adev: amdgpu_device pointer
- * @ih: amdgpu_ih_ring pointet
+ * @ih: amdgpu_ih_ring pointer
 * @enable: true - enable the interrupts, false - disable the interrupts
 *
 * Toggle the interrupt ring buffer (IH_V6_0)
@ -381,6 +381,7 @@ static void ih_v6_0_irq_disable(struct amdgpu_device *adev)
 * ih_v6_0_get_wptr - get the IH ring buffer wptr
 *
 * @adev: amdgpu_device pointer
+ * @ih: amdgpu_ih_ring pointer
 *
 * Get the IH ring buffer wptr from either the register
 * or the writeback memory buffer.  Also check for
@ -425,6 +426,7 @@ out:
 * ih_v6_0_irq_rearm - rearm IRQ if lost
 *
 * @adev: amdgpu_device pointer
+ * @ih: amdgpu_ih_ring pointer
 *
 */
 static void ih_v6_0_irq_rearm(struct amdgpu_device *adev,
@ -450,6 +452,7 @@ static void ih_v6_0_irq_rearm(struct amdgpu_device *adev,
 * ih_v6_0_set_rptr - set the IH ring buffer rptr
 *
 * @adev: amdgpu_device pointer
+ * @ih: amdgpu_ih_ring pointer
 *
 * Set the IH ring buffer rptr.
 */
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
@ -28,6 +28,7 @@
 #include "soc15d.h"
 #include "jpeg_v2_0.h"
 #include "jpeg_v4_0.h"
+#include "mmsch_v4_0.h"

 #include "vcn/vcn_4_0_0_offset.h"
 #include "vcn/vcn_4_0_0_sh_mask.h"
@ -35,12 +36,15 @@

 #define regUVD_JPEG_PITCH_INTERNAL_OFFSET                  0x401f

+static int jpeg_v4_0_start_sriov(struct amdgpu_device *adev);
 static void jpeg_v4_0_set_dec_ring_funcs(struct amdgpu_device *adev);
 static void jpeg_v4_0_set_irq_funcs(struct amdgpu_device *adev);
 static int jpeg_v4_0_set_powergating_state(void *handle,
 				enum amd_powergating_state state);
 static void jpeg_v4_0_set_ras_funcs(struct amdgpu_device *adev);

+static void jpeg_v4_0_dec_ring_set_wptr(struct amdgpu_ring *ring);
+
 /**
 * jpeg_v4_0_early_init - set function pointers
 *
@ -103,7 +107,8 @@ static int jpeg_v4_0_sw_init(void *handle)

 	ring = &adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = true;
-	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1;
+	ring->doorbell_index = amdgpu_sriov_vf(adev) ? (((adev->doorbell_index.vcn.vcn_ring0_1) << 1) + 4) : ((adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1);
+
 	sprintf(ring->name, "jpeg_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
 			     AMDGPU_RING_PRIO_DEFAULT, NULL);
@ -153,16 +158,26 @@ static int jpeg_v4_0_hw_init(void *handle)
 	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
 	int r;

-	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-					(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);
+	if (amdgpu_sriov_vf(adev)) {
+		r = jpeg_v4_0_start_sriov(adev);
+		if (r)
+			return r;
+		ring->wptr = 0;
+		ring->wptr_old = 0;
+		jpeg_v4_0_dec_ring_set_wptr(ring);
+		ring->sched.ready = true;
+	} else {
+		adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+						(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);

-	WREG32_SOC15(VCN, 0, regVCN_JPEG_DB_CTRL,
-		ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
-		VCN_JPEG_DB_CTRL__EN_MASK);
+		WREG32_SOC15(VCN, 0, regVCN_JPEG_DB_CTRL,
+			ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
+			VCN_JPEG_DB_CTRL__EN_MASK);

-	r = amdgpu_ring_test_helper(ring);
-	if (r)
-		return r;
+		r = amdgpu_ring_test_helper(ring);
+		if (r)
+			return r;
+	}

 	DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n");

@ -181,11 +196,11 @@ static int jpeg_v4_0_hw_fini(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

 	cancel_delayed_work_sync(&adev->vcn.idle_work);
-
-	if (adev->jpeg.cur_state != AMD_PG_STATE_GATE &&
-	      RREG32_SOC15(JPEG, 0, regUVD_JRBC_STATUS))
-		jpeg_v4_0_set_powergating_state(adev, AMD_PG_STATE_GATE);
-
+	if (!amdgpu_sriov_vf(adev)) {
+		if (adev->jpeg.cur_state != AMD_PG_STATE_GATE &&
+			RREG32_SOC15(JPEG, 0, regUVD_JRBC_STATUS))
+			jpeg_v4_0_set_powergating_state(adev, AMD_PG_STATE_GATE);
+	}
 	amdgpu_irq_put(adev, &adev->jpeg.inst->irq, 0);

 	return 0;
@ -390,6 +405,120 @@ static int jpeg_v4_0_start(struct amdgpu_device *adev)
 	return 0;
 }

+static int jpeg_v4_0_start_sriov(struct amdgpu_device *adev)
+{
+	struct amdgpu_ring *ring;
+	uint64_t ctx_addr;
+	uint32_t param, resp, expected;
+	uint32_t tmp, timeout;
+
+	struct amdgpu_mm_table *table = &adev->virt.mm_table;
+	uint32_t *table_loc;
+	uint32_t table_size;
+	uint32_t size, size_dw;
+	uint32_t init_status;
+
+	struct mmsch_v4_0_cmd_direct_write
+		direct_wt = { {0} };
+	struct mmsch_v4_0_cmd_end end = { {0} };
+	struct mmsch_v4_0_init_header header;
+
+	direct_wt.cmd_header.command_type =
+		MMSCH_COMMAND__DIRECT_REG_WRITE;
+	end.cmd_header.command_type =
+		MMSCH_COMMAND__END;
+
+	header.version = MMSCH_VERSION;
+	header.total_size = sizeof(struct mmsch_v4_0_init_header) >> 2;
+
+	header.jpegdec.init_status = 0;
+	header.jpegdec.table_offset = 0;
+	header.jpegdec.table_size = 0;
+
+	table_loc = (uint32_t *)table->cpu_addr;
+	table_loc += header.total_size;
+
+	table_size = 0;
+
+	ring = &adev->jpeg.inst->ring_dec;
+
+	MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(JPEG, 0,
+		regUVD_LMI_JRBC_RB_64BIT_BAR_LOW),
+		lower_32_bits(ring->gpu_addr));
+	MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(JPEG, 0,
+		regUVD_LMI_JRBC_RB_64BIT_BAR_HIGH),
+		upper_32_bits(ring->gpu_addr));
+	MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(JPEG, 0,
+		regUVD_JRBC_RB_SIZE), ring->ring_size / 4);
+
+	/* add end packet */
+	MMSCH_V4_0_INSERT_END();
+
+	/* refine header */
+	header.jpegdec.init_status = 0;
+	header.jpegdec.table_offset = header.total_size;
+	header.jpegdec.table_size = table_size;
+	header.total_size += table_size;
+
+	/* Update init table header in memory */
+	size = sizeof(struct mmsch_v4_0_init_header);
+	table_loc = (uint32_t *)table->cpu_addr;
+	memcpy((void *)table_loc, &header, size);
+
+	/* message MMSCH (in VCN[0]) to initialize this client
+	 * 1, write to mmsch_vf_ctx_addr_lo/hi register with GPU mc addr
+	 * of memory descriptor location
+	 */
+	ctx_addr = table->gpu_addr;
+	WREG32_SOC15(VCN, 0, regMMSCH_VF_CTX_ADDR_LO, lower_32_bits(ctx_addr));
+	WREG32_SOC15(VCN, 0, regMMSCH_VF_CTX_ADDR_HI, upper_32_bits(ctx_addr));
+
+	/* 2, update vmid of descriptor */
+	tmp = RREG32_SOC15(VCN, 0, regMMSCH_VF_VMID);
+	tmp &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK;
+	/* use domain0 for MM scheduler */
+	tmp |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT);
+	WREG32_SOC15(VCN, 0, regMMSCH_VF_VMID, tmp);
+
+	/* 3, notify mmsch about the size of this descriptor */
+	size = header.total_size;
+	WREG32_SOC15(VCN, 0, regMMSCH_VF_CTX_SIZE, size);
+
+	/* 4, set resp to zero */
+	WREG32_SOC15(VCN, 0, regMMSCH_VF_MAILBOX_RESP, 0);
+
+	/* 5, kick off the initialization and wait until
+	 * MMSCH_VF_MAILBOX_RESP becomes non-zero
+	 */
+	param = 0x00000001;
+	WREG32_SOC15(VCN, 0, regMMSCH_VF_MAILBOX_HOST, param);
+	tmp = 0;
+	timeout = 1000;
+	resp = 0;
+	expected = MMSCH_VF_MAILBOX_RESP__OK;
+	init_status = ((struct mmsch_v4_0_init_header *)(table_loc))->jpegdec.init_status;
+	while (resp != expected) {
+		resp = RREG32_SOC15(VCN, 0, regMMSCH_VF_MAILBOX_RESP);
+
+		if (resp != 0)
+			break;
+		udelay(10);
+		tmp = tmp + 10;
+		if (tmp >= timeout) {
+			DRM_ERROR("failed to init MMSCH. TIME-OUT after %d usec"\
+				" waiting for regMMSCH_VF_MAILBOX_RESP "\
+				"(expected=0x%08x, readback=0x%08x)\n",
+				tmp, expected, resp);
+			return -EBUSY;
+		}
+	}
+	if (resp != expected && resp != MMSCH_VF_MAILBOX_RESP__INCOMPLETE && init_status != MMSCH_VF_ENGINE_STATUS__PASS)
+		DRM_ERROR("MMSCH init status is incorrect! readback=0x%08x, header init status for jpeg: %x\n", resp, init_status);
+
+	return 0;
+
+}
+
 /**
 * jpeg_v4_0_stop - stop JPEG block
 *
@ -513,6 +642,11 @@ static int jpeg_v4_0_set_powergating_state(void *handle,
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int ret;

+	if (amdgpu_sriov_vf(adev)) {
+		adev->jpeg.cur_state = AMD_PG_STATE_UNGATE;
+		return 0;
+	}
+
 	if (state == adev->jpeg.cur_state)
 		return 0;

--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@ -0,0 +1,477 @@
+/*
+ * Copyright 2022 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 "amdgpu.h"
+#include "mmhub_v1_8.h"
+
+#include "mmhub/mmhub_1_8_0_offset.h"
+#include "mmhub/mmhub_1_8_0_sh_mask.h"
+#include "vega10_enum.h"
+
+#include "soc15_common.h"
+#include "soc15.h"
+
+#define regVM_L2_CNTL3_DEFAULT	0x80100007
+#define regVM_L2_CNTL4_DEFAULT	0x000000c1
+
+static u64 mmhub_v1_8_get_fb_location(struct amdgpu_device *adev)
+{
+	u64 base = RREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_BASE);
+	u64 top = RREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_TOP);
+
+	base &= MC_VM_FB_LOCATION_BASE__FB_BASE_MASK;
+	base <<= 24;
+
+	top &= MC_VM_FB_LOCATION_TOP__FB_TOP_MASK;
+	top <<= 24;
+
+	adev->gmc.fb_start = base;
+	adev->gmc.fb_end = top;
+
+	return base;
+}
+
+static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+				uint64_t page_table_base)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+
+	WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+			    hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base));
+
+	WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+			    hub->ctx_addr_distance * vmid, upper_32_bits(page_table_base));
+}
+
+static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev)
+{
+	uint64_t pt_base;
+
+	if (adev->gmc.pdb0_bo)
+		pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
+	else
+		pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+	mmhub_v1_8_setup_vm_pt_regs(adev, 0, pt_base);
+
+	/* If use GART for FB translation, vmid0 page table covers both
+	 * vram and system memory (gart)
+	 */
+	if (adev->gmc.pdb0_bo) {
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+			     (u32)(adev->gmc.fb_start >> 12));
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+			     (u32)(adev->gmc.fb_start >> 44));
+
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+			     (u32)(adev->gmc.gart_end >> 12));
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+			     (u32)(adev->gmc.gart_end >> 44));
+
+	} else {
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+			     (u32)(adev->gmc.gart_start >> 12));
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+			     (u32)(adev->gmc.gart_start >> 44));
+
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+			     (u32)(adev->gmc.gart_end >> 12));
+		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+			     (u32)(adev->gmc.gart_end >> 44));
+	}
+}
+
+static void mmhub_v1_8_init_system_aperture_regs(struct amdgpu_device *adev)
+{
+	uint64_t value;
+	uint32_t tmp;
+
+	/* Program the AGP BAR */
+	WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BASE, 0);
+	WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
+	WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24);
+
+	/* Program the system aperture low logical page number. */
+	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
+		     min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
+
+	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
+		     max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
+
+	/* In the case squeezing vram into GART aperture, we don't use
+	 * FB aperture and AGP aperture. Disable them.
+	 */
+	if (adev->gmc.pdb0_bo) {
+		WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BOT, 0xFFFFFF);
+		WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_TOP, 0);
+		WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_TOP, 0);
+		WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF);
+		WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF);
+		WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
+	}
+	if (amdgpu_sriov_vf(adev))
+		return;
+
+	/* Set default page address. */
+	value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
+	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
+		     (u32)(value >> 12));
+	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
+		     (u32)(value >> 44));
+
+	/* Program "protection fault". */
+	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
+		     (u32)(adev->dummy_page_addr >> 12));
+	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
+		     (u32)((u64)adev->dummy_page_addr >> 44));
+
+	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL2);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2,
+			    ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
+	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+}
+
+static void mmhub_v1_8_init_tlb_regs(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+
+	/* Setup TLB control */
+	tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL);
+
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+			    ENABLE_ADVANCED_DRIVER_MODEL, 1);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+			    SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+			    MTYPE, MTYPE_UC);/* XXX for emulation. */
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
+
+	WREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
+}
+
+static void mmhub_v1_8_init_cache_regs(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+
+	if (amdgpu_sriov_vf(adev))
+		return;
+
+	/* Setup L2 cache */
+	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1);
+	/* XXX for emulation, Refer to closed source code.*/
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE,
+			    0);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL, tmp);
+
+	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL2);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL2, tmp);
+
+	tmp = regVM_L2_CNTL3_DEFAULT;
+	if (adev->gmc.translate_further) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+	} else {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+	}
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL3, tmp);
+
+	tmp = regVM_L2_CNTL4_DEFAULT;
+	if (adev->gmc.xgmi.connected_to_cpu) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+				    VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+				    VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
+	} else {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+				    VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+				    VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
+	}
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL4, tmp);
+}
+
+static void mmhub_v1_8_enable_system_domain(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+
+	tmp = RREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
+			adev->gmc.vmid0_page_table_depth);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE,
+			adev->gmc.vmid0_page_table_block_size);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL,
+			    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+	WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_CNTL, tmp);
+}
+
+static void mmhub_v1_8_disable_identity_aperture(struct amdgpu_device *adev)
+{
+	if (amdgpu_sriov_vf(adev))
+		return;
+
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32, 0xFFFFFFFF);
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32, 0x0000000F);
+
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32, 0);
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32, 0);
+
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0);
+	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0);
+}
+
+static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	unsigned num_level, block_size;
+	uint32_t tmp;
+	int i;
+
+	num_level = adev->vm_manager.num_level;
+	block_size = adev->vm_manager.block_size;
+	if (adev->gmc.translate_further)
+		num_level -= 1;
+	else
+		block_size -= 9;
+
+	for (i = 0; i <= 14; i++) {
+		tmp = RREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_CNTL, i);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
+				    num_level);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
+				    1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    PAGE_TABLE_BLOCK_SIZE,
+				    block_size);
+		/* On Aldebaran, XNACK can be enabled in the SQ per-process.
+		 * Retry faults need to be enabled for that to work.
+		 */
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
+				    1);
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_CNTL,
+				    i * hub->ctx_distance, tmp);
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
+				    i * hub->ctx_addr_distance, 0);
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
+				    i * hub->ctx_addr_distance, 0);
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
+				    i * hub->ctx_addr_distance,
+				    lower_32_bits(adev->vm_manager.max_pfn - 1));
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
+				    i * hub->ctx_addr_distance,
+				    upper_32_bits(adev->vm_manager.max_pfn - 1));
+	}
+}
+
+static void mmhub_v1_8_program_invalidation(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	unsigned i;
+
+	for (i = 0; i < 18; ++i) {
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
+				    i * hub->eng_addr_distance, 0xffffffff);
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
+				    i * hub->eng_addr_distance, 0x1f);
+	}
+}
+
+static int mmhub_v1_8_gart_enable(struct amdgpu_device *adev)
+{
+	if (amdgpu_sriov_vf(adev)) {
+		/*
+		 * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are
+		 * VF copy registers so vbios post doesn't program them, for
+		 * SRIOV driver need to program them
+		 */
+		WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_BASE,
+			     adev->gmc.vram_start >> 24);
+		WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_TOP,
+			     adev->gmc.vram_end >> 24);
+	}
+
+	/* GART Enable. */
+	mmhub_v1_8_init_gart_aperture_regs(adev);
+	mmhub_v1_8_init_system_aperture_regs(adev);
+	mmhub_v1_8_init_tlb_regs(adev);
+	mmhub_v1_8_init_cache_regs(adev);
+
+	mmhub_v1_8_enable_system_domain(adev);
+	mmhub_v1_8_disable_identity_aperture(adev);
+	mmhub_v1_8_setup_vmid_config(adev);
+	mmhub_v1_8_program_invalidation(adev);
+
+	return 0;
+}
+
+static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	u32 tmp;
+	u32 i;
+
+	/* Disable all tables */
+	for (i = 0; i < 16; i++)
+		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL,
+				    i * hub->ctx_distance, 0);
+
+	/* Setup TLB control */
+	tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0);
+	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+			    ENABLE_ADVANCED_DRIVER_MODEL, 0);
+	WREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
+
+	if (!amdgpu_sriov_vf(adev)) {
+		/* Setup L2 cache */
+		tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0);
+		WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL, tmp);
+		WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL3, 0);
+	}
+}
+
+/**
+ * mmhub_v1_8_set_fault_enable_default - update GART/VM fault handling
+ *
+ * @adev: amdgpu_device pointer
+ * @value: true redirects VM faults to the default page
+ */
+static void mmhub_v1_8_set_fault_enable_default(struct amdgpu_device *adev, bool value)
+{
+	u32 tmp;
+
+	if (amdgpu_sriov_vf(adev))
+		return;
+
+	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
+			    value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	if (!value) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				    CRASH_ON_NO_RETRY_FAULT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				    CRASH_ON_RETRY_FAULT, 1);
+	}
+
+	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL, tmp);
+}
+
+static void mmhub_v1_8_init(struct amdgpu_device *adev)
+{
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+
+	hub->ctx0_ptb_addr_lo32 =
+		SOC15_REG_OFFSET(MMHUB, 0,
+				 regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
+	hub->ctx0_ptb_addr_hi32 =
+		SOC15_REG_OFFSET(MMHUB, 0,
+				 regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
+	hub->vm_inv_eng0_req =
+		SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_REQ);
+	hub->vm_inv_eng0_ack =
+		SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ACK);
+	hub->vm_context0_cntl =
+		SOC15_REG_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL);
+	hub->vm_l2_pro_fault_status =
+		SOC15_REG_OFFSET(MMHUB, 0, regVM_L2_PROTECTION_FAULT_STATUS);
+	hub->vm_l2_pro_fault_cntl =
+		SOC15_REG_OFFSET(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL);
+
+	hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL;
+	hub->ctx_addr_distance = regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
+		regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
+	hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - regVM_INVALIDATE_ENG0_REQ;
+	hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
+		regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+
+}
+
+static int mmhub_v1_8_set_clockgating(struct amdgpu_device *adev,
+				      enum amd_clockgating_state state)
+{
+	return 0;
+}
+
+static void mmhub_v1_8_get_clockgating(struct amdgpu_device *adev, u64 *flags)
+{
+
+}
+
+const struct amdgpu_mmhub_funcs mmhub_v1_8_funcs = {
+	.get_fb_location = mmhub_v1_8_get_fb_location,
+	.init = mmhub_v1_8_init,
+	.gart_enable = mmhub_v1_8_gart_enable,
+	.set_fault_enable_default = mmhub_v1_8_set_fault_enable_default,
+	.gart_disable = mmhub_v1_8_gart_disable,
+	.setup_vm_pt_regs = mmhub_v1_8_setup_vm_pt_regs,
+	.set_clockgating = mmhub_v1_8_set_clockgating,
+	.get_clockgating = mmhub_v1_8_get_clockgating,
+};
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h
@ -0,0 +1,28 @@
+/*
+ * Copyright 2022 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.
+ *
+ */
+#ifndef __MMHUB_V1_8_H__
+#define __MMHUB_V1_8_H__
+
+extern const struct amdgpu_mmhub_funcs mmhub_v1_8_funcs;
+
+#endif
--- a/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h
@ -38,6 +38,11 @@
 #define MMSCH_VF_MAILBOX_RESP__OK 0x1
 #define MMSCH_VF_MAILBOX_RESP__INCOMPLETE 0x2

+#define MMSCH_VF_ENGINE_STATUS__PASS 0x1
+
+#define MMSCH_VF_MAILBOX_RESP__OK 0x1
+#define MMSCH_VF_MAILBOX_RESP__INCOMPLETE 0x2
+
 enum mmsch_v4_0_command_type {
 	MMSCH_COMMAND__DIRECT_REG_WRITE = 0,
 	MMSCH_COMMAND__DIRECT_REG_POLLING = 2,
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.c
@ -26,6 +26,7 @@

 #include "nbio/nbio_4_3_0_offset.h"
 #include "nbio/nbio_4_3_0_sh_mask.h"
+#include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"
 #include <uapi/linux/kfd_ioctl.h>

 static void nbio_v4_3_remap_hdp_registers(struct amdgpu_device *adev)
@ -538,3 +539,81 @@ const struct amdgpu_nbio_funcs nbio_v4_3_sriov_funcs = {
 	.remap_hdp_registers = nbio_v4_3_remap_hdp_registers,
 	.get_rom_offset = nbio_v4_3_get_rom_offset,
 };
+
+static int nbio_v4_3_set_ras_err_event_athub_irq_state(struct amdgpu_device *adev,
+						       struct amdgpu_irq_src *src,
+						       unsigned type,
+						       enum amdgpu_interrupt_state state)
+{
+	/* The ras_controller_irq enablement should be done in psp bl when it
+	 * tries to enable ras feature. Driver only need to set the correct interrupt
+	 * vector for bare-metal and sriov use case respectively
+	 */
+	uint32_t bif_doorbell_int_cntl;
+
+	bif_doorbell_int_cntl = RREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL);
+	bif_doorbell_int_cntl = REG_SET_FIELD(bif_doorbell_int_cntl,
+					      BIF_BX0_BIF_DOORBELL_INT_CNTL,
+					      RAS_ATHUB_ERR_EVENT_INTERRUPT_DISABLE,
+					      (state == AMDGPU_IRQ_STATE_ENABLE) ? 0 : 1);
+	WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL, bif_doorbell_int_cntl);
+
+	return 0;
+}
+
+static int nbio_v4_3_process_err_event_athub_irq(struct amdgpu_device *adev,
+						 struct amdgpu_irq_src *source,
+						 struct amdgpu_iv_entry *entry)
+{
+	/* By design, the ih cookie for err_event_athub_irq should be written
+	 * to bif ring. since bif ring is not enabled, just leave process callback
+	 * as a dummy one.
+	 */
+	return 0;
+}
+
+static const struct amdgpu_irq_src_funcs nbio_v4_3_ras_err_event_athub_irq_funcs = {
+	.set = nbio_v4_3_set_ras_err_event_athub_irq_state,
+	.process = nbio_v4_3_process_err_event_athub_irq,
+};
+
+static void nbio_v4_3_handle_ras_err_event_athub_intr_no_bifring(struct amdgpu_device *adev)
+{
+	uint32_t bif_doorbell_int_cntl;
+
+	bif_doorbell_int_cntl = RREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL);
+	if (REG_GET_FIELD(bif_doorbell_int_cntl,
+			  BIF_DOORBELL_INT_CNTL,
+			  RAS_ATHUB_ERR_EVENT_INTERRUPT_STATUS)) {
+		/* driver has to clear the interrupt status when bif ring is disabled */
+		bif_doorbell_int_cntl = REG_SET_FIELD(bif_doorbell_int_cntl,
+						BIF_DOORBELL_INT_CNTL,
+						RAS_ATHUB_ERR_EVENT_INTERRUPT_CLEAR, 1);
+		WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL, bif_doorbell_int_cntl);
+		amdgpu_ras_global_ras_isr(adev);
+	}
+}
+
+static int nbio_v4_3_init_ras_err_event_athub_interrupt(struct amdgpu_device *adev)
+{
+
+	int r;
+
+	/* init the irq funcs */
+	adev->nbio.ras_err_event_athub_irq.funcs =
+		&nbio_v4_3_ras_err_event_athub_irq_funcs;
+	adev->nbio.ras_err_event_athub_irq.num_types = 1;
+
+	/* register ras err event athub interrupt
+	 * nbio v4_3 uses the same irq source as nbio v7_4 */
+	r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_BIF,
+			      NBIF_7_4__SRCID__ERREVENT_ATHUB_INTERRUPT,
+			      &adev->nbio.ras_err_event_athub_irq);
+
+	return r;
+}
+
+struct amdgpu_nbio_ras nbio_v4_3_ras = {
+	.handle_ras_err_event_athub_intr_no_bifring = nbio_v4_3_handle_ras_err_event_athub_intr_no_bifring,
+	.init_ras_err_event_athub_interrupt = nbio_v4_3_init_ras_err_event_athub_interrupt,
+};
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.h
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.h
@ -29,5 +29,6 @@
 extern const struct nbio_hdp_flush_reg nbio_v4_3_hdp_flush_reg;
 extern const struct amdgpu_nbio_funcs nbio_v4_3_funcs;
 extern const struct amdgpu_nbio_funcs nbio_v4_3_sriov_funcs;
+extern struct amdgpu_nbio_ras nbio_v4_3_ras;

 #endif
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@ -0,0 +1,369 @@
+/*
+ * Copyright 2022 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 "amdgpu.h"
+#include "amdgpu_atombios.h"
+#include "nbio_v7_9.h"
+#include "amdgpu_ras.h"
+
+#include "nbio/nbio_7_9_0_offset.h"
+#include "nbio/nbio_7_9_0_sh_mask.h"
+#include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"
+#include <uapi/linux/kfd_ioctl.h>
+
+static void nbio_v7_9_remap_hdp_registers(struct amdgpu_device *adev)
+{
+	WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_MEM_FLUSH_CNTL,
+		adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL);
+	WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_REG_FLUSH_CNTL,
+		adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_REG_FLUSH_CNTL);
+}
+
+static u32 nbio_v7_9_get_rev_id(struct amdgpu_device *adev)
+{
+	u32 tmp;
+
+	tmp = RREG32_SOC15(NBIO, 0, regRCC_STRAP0_RCC_DEV0_EPF0_STRAP0);
+	tmp = REG_GET_FIELD(tmp, RCC_STRAP0_RCC_DEV0_EPF0_STRAP0, STRAP_ATI_REV_ID_DEV0_F0);
+
+	return tmp;
+}
+
+static void nbio_v7_9_mc_access_enable(struct amdgpu_device *adev, bool enable)
+{
+	if (enable)
+		WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_FB_EN,
+			BIF_BX0_BIF_FB_EN__FB_READ_EN_MASK | BIF_BX0_BIF_FB_EN__FB_WRITE_EN_MASK);
+	else
+		WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_FB_EN, 0);
+}
+
+static u32 nbio_v7_9_get_memsize(struct amdgpu_device *adev)
+{
+	return RREG32_SOC15(NBIO, 0, regRCC_DEV0_EPF0_RCC_CONFIG_MEMSIZE);
+}
+
+static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instance,
+			bool use_doorbell, int doorbell_index, int doorbell_size)
+{
+	u32 doorbell_range = 0, doorbell_ctrl = 0;
+
+	doorbell_range =
+		REG_SET_FIELD(doorbell_range, DOORBELL0_CTRL_ENTRY_0,
+			BIF_DOORBELL0_RANGE_OFFSET_ENTRY, doorbell_index);
+	doorbell_range =
+		REG_SET_FIELD(doorbell_range, DOORBELL0_CTRL_ENTRY_0,
+			BIF_DOORBELL0_RANGE_SIZE_ENTRY, doorbell_size);
+	doorbell_ctrl =
+		REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL,
+			S2A_DOORBELL_PORT1_ENABLE, 1);
+	doorbell_ctrl =
+		REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL,
+			S2A_DOORBELL_PORT1_RANGE_SIZE, doorbell_size);
+
+	switch (instance) {
+	case 0:
+		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1, doorbell_range);
+
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWID, 0xe);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_RANGE_OFFSET, 0xe);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
+					0x1);
+		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_1_CTRL, doorbell_ctrl);
+		break;
+	case 1:
+		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2, doorbell_range);
+
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWID, 0x8);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_RANGE_OFFSET, 0x8);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
+					0x2);
+		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_2_CTRL, doorbell_ctrl);
+		break;
+	case 2:
+		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3, doorbell_range);
+
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWID, 0x9);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_RANGE_OFFSET, 0x9);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
+					0x8);
+		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_5_CTRL, doorbell_ctrl);
+		break;
+	case 3:
+		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4, doorbell_range);
+
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWID, 0xa);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_RANGE_OFFSET, 0xa);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+					S2A_DOORBELL_ENTRY_1_CTRL,
+					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
+					0x9);
+		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_6_CTRL, doorbell_ctrl);
+		break;
+	default:
+		break;
+	};
+
+	return;
+}
+
+static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_doorbell,
+					 int doorbell_index, int instance)
+{
+	u32 doorbell_range = 0, doorbell_ctrl = 0;
+
+	if (use_doorbell) {
+		doorbell_range = REG_SET_FIELD(doorbell_range,
+				DOORBELL0_CTRL_ENTRY_0,
+				BIF_DOORBELL0_RANGE_OFFSET_ENTRY,
+				doorbell_index);
+		doorbell_range = REG_SET_FIELD(doorbell_range,
+				DOORBELL0_CTRL_ENTRY_0,
+				BIF_DOORBELL0_RANGE_SIZE_ENTRY,
+				0x8);
+
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_ENABLE, 1);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_AWID, 0x4);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_RANGE_OFFSET, 0x4);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x8);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x4);
+	} else {
+		doorbell_range = REG_SET_FIELD(doorbell_range,
+				DOORBELL0_CTRL_ENTRY_0,
+				BIF_DOORBELL0_RANGE_SIZE_ENTRY, 0);
+		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_RANGE_SIZE, 0);
+	}
+
+	WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17, doorbell_range);
+	WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL, doorbell_ctrl);
+}
+
+static void nbio_v7_9_enable_doorbell_aperture(struct amdgpu_device *adev,
+					       bool enable)
+{
+	/* Enable to allow doorbell pass thru on pre-silicon bare-metal */
+	WREG32_SOC15(NBIO, 0, regBIFC_DOORBELL_ACCESS_EN_PF, 0xfffff);
+	WREG32_FIELD15_PREREG(NBIO, 0, RCC_DEV0_EPF0_RCC_DOORBELL_APER_EN,
+			BIF_DOORBELL_APER_EN, enable ? 1 : 0);
+}
+
+static void nbio_v7_9_enable_doorbell_selfring_aperture(struct amdgpu_device *adev,
+							bool enable)
+{
+	u32 tmp = 0;
+
+	if (enable) {
+		tmp = REG_SET_FIELD(tmp, BIF_BX_PF0_DOORBELL_SELFRING_GPA_APER_CNTL,
+				    DOORBELL_SELFRING_GPA_APER_EN, 1) |
+		      REG_SET_FIELD(tmp, BIF_BX_PF0_DOORBELL_SELFRING_GPA_APER_CNTL,
+				    DOORBELL_SELFRING_GPA_APER_MODE, 1) |
+		      REG_SET_FIELD(tmp, BIF_BX_PF0_DOORBELL_SELFRING_GPA_APER_CNTL,
+				    DOORBELL_SELFRING_GPA_APER_SIZE, 0);
+
+		WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_DOORBELL_SELFRING_GPA_APER_BASE_LOW,
+			     lower_32_bits(adev->doorbell.base));
+		WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_DOORBELL_SELFRING_GPA_APER_BASE_HIGH,
+			     upper_32_bits(adev->doorbell.base));
+	}
+
+	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_DOORBELL_SELFRING_GPA_APER_CNTL, tmp);
+}
+
+static void nbio_v7_9_ih_doorbell_range(struct amdgpu_device *adev,
+					bool use_doorbell, int doorbell_index)
+{
+	u32 ih_doorbell_range = 0, ih_doorbell_ctrl = 0;
+
+	if (use_doorbell) {
+		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range,
+				DOORBELL0_CTRL_ENTRY_0,
+				BIF_DOORBELL0_RANGE_OFFSET_ENTRY,
+				doorbell_index);
+		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range,
+				DOORBELL0_CTRL_ENTRY_0,
+				BIF_DOORBELL0_RANGE_SIZE_ENTRY,
+				0x4);
+
+		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_ENABLE, 1);
+		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_AWID, 0);
+		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_RANGE_OFFSET, 0);
+		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x4);
+		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0);
+	} else {
+		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range,
+				DOORBELL0_CTRL_ENTRY_0,
+				BIF_DOORBELL0_RANGE_SIZE_ENTRY, 0);
+		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
+				S2A_DOORBELL_ENTRY_1_CTRL,
+				S2A_DOORBELL_PORT1_RANGE_SIZE, 0);
+	}
+
+	WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_0, ih_doorbell_range);
+	WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_3_CTRL, ih_doorbell_ctrl);
+}
+
+
+static void nbio_v7_9_update_medium_grain_clock_gating(struct amdgpu_device *adev,
+						       bool enable)
+{
+}
+
+static void nbio_v7_9_update_medium_grain_light_sleep(struct amdgpu_device *adev,
+						      bool enable)
+{
+}
+
+static void nbio_v7_9_get_clockgating_state(struct amdgpu_device *adev,
+					    u64 *flags)
+{
+}
+
+static void nbio_v7_9_ih_control(struct amdgpu_device *adev)
+{
+	u32 interrupt_cntl;
+
+	/* setup interrupt control */
+	WREG32_SOC15(NBIO, 0, regBIF_BX0_INTERRUPT_CNTL2, adev->dummy_page_addr >> 8);
+	interrupt_cntl = RREG32_SOC15(NBIO, 0, regBIF_BX0_INTERRUPT_CNTL);
+	/* INTERRUPT_CNTL__IH_DUMMY_RD_OVERRIDE_MASK=0 - dummy read disabled with msi, enabled without msi
+	 * INTERRUPT_CNTL__IH_DUMMY_RD_OVERRIDE_MASK=1 - dummy read controlled by IH_DUMMY_RD_EN
+	 */
+	interrupt_cntl =
+		REG_SET_FIELD(interrupt_cntl, BIF_BX0_INTERRUPT_CNTL, IH_DUMMY_RD_OVERRIDE, 0);
+	/* INTERRUPT_CNTL__IH_REQ_NONSNOOP_EN_MASK=1 if ring is in non-cacheable memory, e.g., vram */
+	interrupt_cntl =
+		REG_SET_FIELD(interrupt_cntl, BIF_BX0_INTERRUPT_CNTL, IH_REQ_NONSNOOP_EN, 0);
+	WREG32_SOC15(NBIO, 0, regBIF_BX0_INTERRUPT_CNTL, interrupt_cntl);
+}
+
+static u32 nbio_v7_9_get_hdp_flush_req_offset(struct amdgpu_device *adev)
+{
+	return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX_PF0_GPU_HDP_FLUSH_REQ);
+}
+
+static u32 nbio_v7_9_get_hdp_flush_done_offset(struct amdgpu_device *adev)
+{
+	return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX_PF0_GPU_HDP_FLUSH_DONE);
+}
+
+static u32 nbio_v7_9_get_pcie_index_offset(struct amdgpu_device *adev)
+{
+	return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX0_PCIE_INDEX2);
+}
+
+static u32 nbio_v7_9_get_pcie_data_offset(struct amdgpu_device *adev)
+{
+	return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX0_PCIE_DATA2);
+}
+
+const struct nbio_hdp_flush_reg nbio_v7_9_hdp_flush_reg = {
+	.ref_and_mask_cp0 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP0_MASK,
+	.ref_and_mask_cp1 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP1_MASK,
+	.ref_and_mask_cp2 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP2_MASK,
+	.ref_and_mask_cp3 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP3_MASK,
+	.ref_and_mask_cp4 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP4_MASK,
+	.ref_and_mask_cp5 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP5_MASK,
+	.ref_and_mask_cp6 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP6_MASK,
+	.ref_and_mask_cp7 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP7_MASK,
+	.ref_and_mask_cp8 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP8_MASK,
+	.ref_and_mask_cp9 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP9_MASK,
+	.ref_and_mask_sdma0 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__SDMA0_MASK,
+	.ref_and_mask_sdma1 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__SDMA1_MASK,
+	.ref_and_mask_sdma2 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__RSVD_ENG0_MASK,
+	.ref_and_mask_sdma3 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__RSVD_ENG1_MASK,
+	.ref_and_mask_sdma4 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__RSVD_ENG2_MASK,
+	.ref_and_mask_sdma5 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__RSVD_ENG3_MASK,
+	.ref_and_mask_sdma6 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__RSVD_ENG4_MASK,
+	.ref_and_mask_sdma7 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__RSVD_ENG5_MASK,
+};
+
+static void nbio_v7_9_enable_doorbell_interrupt(struct amdgpu_device *adev,
+						bool enable)
+{
+	WREG32_FIELD15_PREREG(NBIO, 0, BIF_BX0_BIF_DOORBELL_INT_CNTL,
+			      DOORBELL_INTERRUPT_DISABLE, enable ? 0 : 1);
+}
+
+const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
+	.get_hdp_flush_req_offset = nbio_v7_9_get_hdp_flush_req_offset,
+	.get_hdp_flush_done_offset = nbio_v7_9_get_hdp_flush_done_offset,
+	.get_pcie_index_offset = nbio_v7_9_get_pcie_index_offset,
+	.get_pcie_data_offset = nbio_v7_9_get_pcie_data_offset,
+	.get_rev_id = nbio_v7_9_get_rev_id,
+	.mc_access_enable = nbio_v7_9_mc_access_enable,
+	.get_memsize = nbio_v7_9_get_memsize,
+	.sdma_doorbell_range = nbio_v7_9_sdma_doorbell_range,
+	.vcn_doorbell_range = nbio_v7_9_vcn_doorbell_range,
+	.enable_doorbell_aperture = nbio_v7_9_enable_doorbell_aperture,
+	.enable_doorbell_selfring_aperture = nbio_v7_9_enable_doorbell_selfring_aperture,
+	.ih_doorbell_range = nbio_v7_9_ih_doorbell_range,
+	.enable_doorbell_interrupt = nbio_v7_9_enable_doorbell_interrupt,
+	.update_medium_grain_clock_gating = nbio_v7_9_update_medium_grain_clock_gating,
+	.update_medium_grain_light_sleep = nbio_v7_9_update_medium_grain_light_sleep,
+	.get_clockgating_state = nbio_v7_9_get_clockgating_state,
+	.ih_control = nbio_v7_9_ih_control,
+	.remap_hdp_registers = nbio_v7_9_remap_hdp_registers,
+};
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.h
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.h
@ -0,0 +1,32 @@
+/*
+ * Copyright 2022 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.
+ *
+ */
+
+#ifndef __NBIO_V7_9_H__
+#define __NBIO_V7_9_H__
+
+#include "soc15_common.h"
+
+extern const struct nbio_hdp_flush_reg nbio_v7_9_hdp_flush_reg;
+extern const struct amdgpu_nbio_funcs nbio_v7_9_funcs;
+
+#endif
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@ -754,6 +754,14 @@ static int soc21_common_late_init(void *handle)
 							     sriov_vcn_4_0_0_video_codecs_decode_array_vcn0,
 							     ARRAY_SIZE(sriov_vcn_4_0_0_video_codecs_decode_array_vcn0));
 		}
+	} else {
+		if (adev->nbio.ras &&
+		    adev->nbio.ras_err_event_athub_irq.funcs)
+			/* don't need to fail gpu late init
+			 * if enabling athub_err_event interrupt failed
+			 * nbio v4_3 only support fatal error hanlding
+			 * just enable the interrupt directly */
+			amdgpu_irq_get(adev, &adev->nbio.ras_err_event_athub_irq, 0);
 	}

 	return 0;
@ -801,8 +809,13 @@ static int soc21_common_hw_fini(void *handle)
 	/* disable the doorbell aperture */
 	soc21_enable_doorbell_aperture(adev, false);

-	if (amdgpu_sriov_vf(adev))
+	if (amdgpu_sriov_vf(adev)) {
 		xgpu_nv_mailbox_put_irq(adev);
+	} else {
+		if (adev->nbio.ras &&
+		    adev->nbio.ras_err_event_athub_irq.funcs)
+			amdgpu_irq_put(adev, &adev->nbio.ras_err_event_athub_irq, 0);
+	}

 	return 0;
 }
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
@ -2936,3 +2936,490 @@ static const uint32_t cwsr_trap_gfx11_hex[] = {
 	0xbf9f0000, 0xbf9f0000,
 	0xbf9f0000, 0x00000000,
 };
+
+static const uint32_t cwsr_trap_gfx9_4_3_hex[] = {
+	0xbf820001, 0xbf8202d6,
+	0xb8f8f802, 0x89788678,
+	0xb8fbf803, 0x866eff78,
+	0x00002000, 0xbf840009,
+	0x866eff6d, 0x00ff0000,
+	0xbf85001a, 0x866eff7b,
+	0x00000400, 0xbf85004d,
+	0xbf8e0010, 0xb8fbf803,
+	0xbf82fffa, 0x866eff7b,
+	0x03c00900, 0xbf850011,
+	0x866eff7b, 0x000071ff,
+	0xbf840008, 0x866fff7b,
+	0x00007080, 0xbf840001,
+	0xbeee1a87, 0xb8eff801,
+	0x8e6e8c6e, 0x866e6f6e,
+	0xbf850006, 0x866eff6d,
+	0x00ff0000, 0xbf850003,
+	0x866eff7b, 0x00000400,
+	0xbf850036, 0xb8faf807,
+	0x867aff7a, 0x001f8000,
+	0x8e7a8b7a, 0x8979ff79,
+	0xfc000000, 0x87797a79,
+	0xba7ff807, 0x00000000,
+	0xb8faf812, 0xb8fbf813,
+	0x8efa887a, 0xc0031bbd,
+	0x00000010, 0xbf8cc07f,
+	0x8e6e976e, 0x8979ff79,
+	0x00800000, 0x87796e79,
+	0xc0071bbd, 0x00000000,
+	0xbf8cc07f, 0xc0071ebd,
+	0x00000008, 0xbf8cc07f,
+	0x86ee6e6e, 0xbf840001,
+	0xbe801d6e, 0x866eff6d,
+	0x01ff0000, 0xbf850005,
+	0x8778ff78, 0x00002000,
+	0x80ec886c, 0x82ed806d,
+	0xbf820005, 0x866eff6d,
+	0x01000000, 0xbf850002,
+	0x806c846c, 0x826d806d,
+	0x866dff6d, 0x0000ffff,
+	0x8f7a8b79, 0x867aff7a,
+	0x001f8000, 0xb97af807,
+	0x86fe7e7e, 0x86ea6a6a,
+	0x8f6e8378, 0xb96ee0c2,
+	0xbf800002, 0xb9780002,
+	0xbe801f6c, 0x866dff6d,
+	0x0000ffff, 0xbefa0080,
+	0xb97a0283, 0xb8faf807,
+	0x867aff7a, 0x001f8000,
+	0x8e7a8b7a, 0x8979ff79,
+	0xfc000000, 0x87797a79,
+	0xba7ff807, 0x00000000,
+	0xbeee007e, 0xbeef007f,
+	0xbefe0180, 0xbf900004,
+	0x877a8478, 0xb97af802,
+	0xbf8e0002, 0xbf88fffe,
+	0xb8fa2985, 0x807a817a,
+	0x8e7a8a7a, 0x8e7a817a,
+	0xb8fb1605, 0x807b817b,
+	0x8e7b867b, 0x807a7b7a,
+	0x807a7e7a, 0x827b807f,
+	0x867bff7b, 0x0000ffff,
+	0xc04b1c3d, 0x00000050,
+	0xbf8cc07f, 0xc04b1d3d,
+	0x00000060, 0xbf8cc07f,
+	0xc0431e7d, 0x00000074,
+	0xbf8cc07f, 0xbef4007e,
+	0x8675ff7f, 0x0000ffff,
+	0x8775ff75, 0x00040000,
+	0xbef60080, 0xbef700ff,
+	0x00807fac, 0xbef1007c,
+	0xbef00080, 0xb8f02985,
+	0x80708170, 0x8e708a70,
+	0x8e708170, 0xb8fa1605,
+	0x807a817a, 0x8e7a867a,
+	0x80707a70, 0xbef60084,
+	0xbef600ff, 0x01000000,
+	0xbefe007c, 0xbefc0070,
+	0xc0611c7a, 0x0000007c,
+	0xbf8cc07f, 0x80708470,
+	0xbefc007e, 0xbefe007c,
+	0xbefc0070, 0xc0611b3a,
+	0x0000007c, 0xbf8cc07f,
+	0x80708470, 0xbefc007e,
+	0xbefe007c, 0xbefc0070,
+	0xc0611b7a, 0x0000007c,
+	0xbf8cc07f, 0x80708470,
+	0xbefc007e, 0xbefe007c,
+	0xbefc0070, 0xc0611bba,
+	0x0000007c, 0xbf8cc07f,
+	0x80708470, 0xbefc007e,
+	0xbefe007c, 0xbefc0070,
+	0xc0611bfa, 0x0000007c,
+	0xbf8cc07f, 0x80708470,
+	0xbefc007e, 0xbefe007c,
+	0xbefc0070, 0xc0611e3a,
+	0x0000007c, 0xbf8cc07f,
+	0x80708470, 0xbefc007e,
+	0xb8fbf803, 0xbefe007c,
+	0xbefc0070, 0xc0611efa,
+	0x0000007c, 0xbf8cc07f,
+	0x80708470, 0xbefc007e,
+	0xbefe007c, 0xbefc0070,
+	0xc0611a3a, 0x0000007c,
+	0xbf8cc07f, 0x80708470,
+	0xbefc007e, 0xbefe007c,
+	0xbefc0070, 0xc0611a7a,
+	0x0000007c, 0xbf8cc07f,
+	0x80708470, 0xbefc007e,
+	0xb8f1f801, 0xbefe007c,
+	0xbefc0070, 0xc0611c7a,
+	0x0000007c, 0xbf8cc07f,
+	0x80708470, 0xbefc007e,
+	0x867aff7f, 0x04000000,
+	0xbeef0080, 0x876f6f7a,
+	0xb8f02985, 0x80708170,
+	0x8e708a70, 0x8e708170,
+	0xb8fb1605, 0x807b817b,
+	0x8e7b847b, 0x8e76827b,
+	0xbef600ff, 0x01000000,
+	0xbef20174, 0x80747074,
+	0x82758075, 0xbefc0080,
+	0xbf800000, 0xbe802b00,
+	0xbe822b02, 0xbe842b04,
+	0xbe862b06, 0xbe882b08,
+	0xbe8a2b0a, 0xbe8c2b0c,
+	0xbe8e2b0e, 0xc06b003a,
+	0x00000000, 0xbf8cc07f,
+	0xc06b013a, 0x00000010,
+	0xbf8cc07f, 0xc06b023a,
+	0x00000020, 0xbf8cc07f,
+	0xc06b033a, 0x00000030,
+	0xbf8cc07f, 0x8074c074,
+	0x82758075, 0x807c907c,
+	0xbf0a7b7c, 0xbf85ffe7,
+	0xbef40172, 0xbef00080,
+	0xbefe00c1, 0xbeff00c1,
+	0xbee80080, 0xbee90080,
+	0xbef600ff, 0x01000000,
+	0x867aff78, 0x00400000,
+	0xbf850003, 0xb8faf803,
+	0x897a7aff, 0x10000000,
+	0xbf85004d, 0xbe840080,
+	0xd2890000, 0x00000900,
+	0x80048104, 0xd2890001,
+	0x00000900, 0x80048104,
+	0xd2890002, 0x00000900,
+	0x80048104, 0xd2890003,
+	0x00000900, 0x80048104,
+	0xc069003a, 0x00000070,
+	0xbf8cc07f, 0x80709070,
+	0xbf06c004, 0xbf84ffee,
+	0xbe840080, 0xd2890000,
+	0x00000901, 0x80048104,
+	0xd2890001, 0x00000901,
+	0x80048104, 0xd2890002,
+	0x00000901, 0x80048104,
+	0xd2890003, 0x00000901,
+	0x80048104, 0xc069003a,
+	0x00000070, 0xbf8cc07f,
+	0x80709070, 0xbf06c004,
+	0xbf84ffee, 0xbe840080,
+	0xd2890000, 0x00000902,
+	0x80048104, 0xd2890001,
+	0x00000902, 0x80048104,
+	0xd2890002, 0x00000902,
+	0x80048104, 0xd2890003,
+	0x00000902, 0x80048104,
+	0xc069003a, 0x00000070,
+	0xbf8cc07f, 0x80709070,
+	0xbf06c004, 0xbf84ffee,
+	0xbe840080, 0xd2890000,
+	0x00000903, 0x80048104,
+	0xd2890001, 0x00000903,
+	0x80048104, 0xd2890002,
+	0x00000903, 0x80048104,
+	0xd2890003, 0x00000903,
+	0x80048104, 0xc069003a,
+	0x00000070, 0xbf8cc07f,
+	0x80709070, 0xbf06c004,
+	0xbf84ffee, 0xbf820008,
+	0xe0724000, 0x701d0000,
+	0xe0724100, 0x701d0100,
+	0xe0724200, 0x701d0200,
+	0xe0724300, 0x701d0300,
+	0xbefe00c1, 0xbeff00c1,
+	0xb8fb4306, 0x867bc17b,
+	0xbf840064, 0xbf8a0000,
+	0x867aff6f, 0x04000000,
+	0xbf840060, 0x8e7b867b,
+	0x8e7b827b, 0xbef6007b,
+	0xb8f02985, 0x80708170,
+	0x8e708a70, 0x8e708170,
+	0xb8fa1605, 0x807a817a,
+	0x8e7a867a, 0x80707a70,
+	0x8070ff70, 0x00000080,
+	0xbef600ff, 0x01000000,
+	0xbefc0080, 0xd28c0002,
+	0x000100c1, 0xd28d0003,
+	0x000204c1, 0x867aff78,
+	0x00400000, 0xbf850003,
+	0xb8faf803, 0x897a7aff,
+	0x10000000, 0xbf850030,
+	0x24040682, 0xd86e4000,
+	0x00000002, 0xbf8cc07f,
+	0xbe840080, 0xd2890000,
+	0x00000900, 0x80048104,
+	0xd2890001, 0x00000900,
+	0x80048104, 0xd2890002,
+	0x00000900, 0x80048104,
+	0xd2890003, 0x00000900,
+	0x80048104, 0xc069003a,
+	0x00000070, 0xbf8cc07f,
+	0x80709070, 0xbf06c004,
+	0xbf84ffee, 0xbe840080,
+	0xd2890000, 0x00000901,
+	0x80048104, 0xd2890001,
+	0x00000901, 0x80048104,
+	0xd2890002, 0x00000901,
+	0x80048104, 0xd2890003,
+	0x00000901, 0x80048104,
+	0xc069003a, 0x00000070,
+	0xbf8cc07f, 0x80709070,
+	0xbf06c004, 0xbf84ffee,
+	0x680404ff, 0x00000200,
+	0xd0c9006a, 0x0000f702,
+	0xbf87ffd2, 0xbf820015,
+	0xd1060002, 0x00011103,
+	0x7e0602ff, 0x00000200,
+	0xbefc00ff, 0x00010000,
+	0xbe800077, 0x8677ff77,
+	0xff7fffff, 0x8777ff77,
+	0x00058000, 0xd8ec0000,
+	0x00000002, 0xbf8cc07f,
+	0xe0765000, 0x701d0002,
+	0x68040702, 0xd0c9006a,
+	0x0000f702, 0xbf87fff7,
+	0xbef70000, 0xbef000ff,
+	0x00000400, 0xbefe00c1,
+	0xbeff00c1, 0xb8fb2b05,
+	0x807b817b, 0x8e7b827b,
+	0xbef600ff, 0x01000000,
+	0xbefc0084, 0xbf0a7b7c,
+	0xbf84006d, 0xbf11017c,
+	0x807bff7b, 0x00001000,
+	0x867aff78, 0x00400000,
+	0xbf850003, 0xb8faf803,
+	0x897a7aff, 0x10000000,
+	0xbf850051, 0xbe840080,
+	0xd2890000, 0x00000900,
+	0x80048104, 0xd2890001,
+	0x00000900, 0x80048104,
+	0xd2890002, 0x00000900,
+	0x80048104, 0xd2890003,
+	0x00000900, 0x80048104,
+	0xc069003a, 0x00000070,
+	0xbf8cc07f, 0x80709070,
+	0xbf06c004, 0xbf84ffee,
+	0xbe840080, 0xd2890000,
+	0x00000901, 0x80048104,
+	0xd2890001, 0x00000901,
+	0x80048104, 0xd2890002,
+	0x00000901, 0x80048104,
+	0xd2890003, 0x00000901,
+	0x80048104, 0xc069003a,
+	0x00000070, 0xbf8cc07f,
+	0x80709070, 0xbf06c004,
+	0xbf84ffee, 0xbe840080,
+	0xd2890000, 0x00000902,
+	0x80048104, 0xd2890001,
+	0x00000902, 0x80048104,
+	0xd2890002, 0x00000902,
+	0x80048104, 0xd2890003,
+	0x00000902, 0x80048104,
+	0xc069003a, 0x00000070,
+	0xbf8cc07f, 0x80709070,
+	0xbf06c004, 0xbf84ffee,
+	0xbe840080, 0xd2890000,
+	0x00000903, 0x80048104,
+	0xd2890001, 0x00000903,
+	0x80048104, 0xd2890002,
+	0x00000903, 0x80048104,
+	0xd2890003, 0x00000903,
+	0x80048104, 0xc069003a,
+	0x00000070, 0xbf8cc07f,
+	0x80709070, 0xbf06c004,
+	0xbf84ffee, 0x807c847c,
+	0xbf0a7b7c, 0xbf85ffb1,
+	0xbf9c0000, 0xbf820012,
+	0x7e000300, 0x7e020301,
+	0x7e040302, 0x7e060303,
+	0xe0724000, 0x701d0000,
+	0xe0724100, 0x701d0100,
+	0xe0724200, 0x701d0200,
+	0xe0724300, 0x701d0300,
+	0x807c847c, 0x8070ff70,
+	0x00000400, 0xbf0a7b7c,
+	0xbf85ffef, 0xbf9c0000,
+	0xb8fb2985, 0x807b817b,
+	0x8e7b837b, 0xb8fa2b05,
+	0x807a817a, 0x8e7a827a,
+	0x80fb7a7b, 0x867b7b7b,
+	0xbf84007a, 0x807bff7b,
+	0x00001000, 0xbefc0080,
+	0xbf11017c, 0x867aff78,
+	0x00400000, 0xbf850003,
+	0xb8faf803, 0x897a7aff,
+	0x10000000, 0xbf850059,
+	0xd3d84000, 0x18000100,
+	0xd3d84001, 0x18000101,
+	0xd3d84002, 0x18000102,
+	0xd3d84003, 0x18000103,
+	0xbe840080, 0xd2890000,
+	0x00000900, 0x80048104,
+	0xd2890001, 0x00000900,
+	0x80048104, 0xd2890002,
+	0x00000900, 0x80048104,
+	0xd2890003, 0x00000900,
+	0x80048104, 0xc069003a,
+	0x00000070, 0xbf8cc07f,
+	0x80709070, 0xbf06c004,
+	0xbf84ffee, 0xbe840080,
+	0xd2890000, 0x00000901,
+	0x80048104, 0xd2890001,
+	0x00000901, 0x80048104,
+	0xd2890002, 0x00000901,
+	0x80048104, 0xd2890003,
+	0x00000901, 0x80048104,
+	0xc069003a, 0x00000070,
+	0xbf8cc07f, 0x80709070,
+	0xbf06c004, 0xbf84ffee,
+	0xbe840080, 0xd2890000,
+	0x00000902, 0x80048104,
+	0xd2890001, 0x00000902,
+	0x80048104, 0xd2890002,
+	0x00000902, 0x80048104,
+	0xd2890003, 0x00000902,
+	0x80048104, 0xc069003a,
+	0x00000070, 0xbf8cc07f,
+	0x80709070, 0xbf06c004,
+	0xbf84ffee, 0xbe840080,
+	0xd2890000, 0x00000903,
+	0x80048104, 0xd2890001,
+	0x00000903, 0x80048104,
+	0xd2890002, 0x00000903,
+	0x80048104, 0xd2890003,
+	0x00000903, 0x80048104,
+	0xc069003a, 0x00000070,
+	0xbf8cc07f, 0x80709070,
+	0xbf06c004, 0xbf84ffee,
+	0x807c847c, 0xbf0a7b7c,
+	0xbf85ffa9, 0xbf9c0000,
+	0xbf820016, 0xd3d84000,
+	0x18000100, 0xd3d84001,
+	0x18000101, 0xd3d84002,
+	0x18000102, 0xd3d84003,
+	0x18000103, 0xe0724000,
+	0x701d0000, 0xe0724100,
+	0x701d0100, 0xe0724200,
+	0x701d0200, 0xe0724300,
+	0x701d0300, 0x807c847c,
+	0x8070ff70, 0x00000400,
+	0xbf0a7b7c, 0xbf85ffeb,
+	0xbf9c0000, 0xbf8200ee,
+	0xbef4007e, 0x8675ff7f,
+	0x0000ffff, 0x8775ff75,
+	0x00040000, 0xbef60080,
+	0xbef700ff, 0x00807fac,
+	0x866eff7f, 0x04000000,
+	0xbf84001f, 0xbefe00c1,
+	0xbeff00c1, 0xb8ef4306,
+	0x866fc16f, 0xbf84001a,
+	0x8e6f866f, 0x8e6f826f,
+	0xbef6006f, 0xb8f82985,
+	0x80788178, 0x8e788a78,
+	0x8e788178, 0xb8ee1605,
+	0x806e816e, 0x8e6e866e,
+	0x80786e78, 0x8078ff78,
+	0x00000080, 0xbef600ff,
+	0x01000000, 0xbefc0080,
+	0xe0510000, 0x781d0000,
+	0xe0510100, 0x781d0000,
+	0x807cff7c, 0x00000200,
+	0x8078ff78, 0x00000200,
+	0xbf0a6f7c, 0xbf85fff6,
+	0xbefe00c1, 0xbeff00c1,
+	0xbef600ff, 0x01000000,
+	0xb8ef2b05, 0x806f816f,
+	0x8e6f826f, 0x806fff6f,
+	0x00008000, 0xbef80080,
+	0xbeee0078, 0x8078ff78,
+	0x00000400, 0xbefc0084,
+	0xbf11087c, 0xe0524000,
+	0x781d0000, 0xe0524100,
+	0x781d0100, 0xe0524200,
+	0x781d0200, 0xe0524300,
+	0x781d0300, 0xbf8c0f70,
+	0x7e000300, 0x7e020301,
+	0x7e040302, 0x7e060303,
+	0x807c847c, 0x8078ff78,
+	0x00000400, 0xbf0a6f7c,
+	0xbf85ffee, 0xb8ef2985,
+	0x806f816f, 0x8e6f836f,
+	0xb8f92b05, 0x80798179,
+	0x8e798279, 0x80ef796f,
+	0x866f6f6f, 0xbf84001a,
+	0x806fff6f, 0x00008000,
+	0xbefc0080, 0xbf11087c,
+	0xe0524000, 0x781d0000,
+	0xe0524100, 0x781d0100,
+	0xe0524200, 0x781d0200,
+	0xe0524300, 0x781d0300,
+	0xbf8c0f70, 0xd3d94000,
+	0x18000100, 0xd3d94001,
+	0x18000101, 0xd3d94002,
+	0x18000102, 0xd3d94003,
+	0x18000103, 0x807c847c,
+	0x8078ff78, 0x00000400,
+	0xbf0a6f7c, 0xbf85ffea,
+	0xbf9c0000, 0xe0524000,
+	0x6e1d0000, 0xe0524100,
+	0x6e1d0100, 0xe0524200,
+	0x6e1d0200, 0xe0524300,
+	0x6e1d0300, 0xbf8c0f70,
+	0xb8f82985, 0x80788178,
+	0x8e788a78, 0x8e788178,
+	0xb8ee1605, 0x806e816e,
+	0x8e6e866e, 0x80786e78,
+	0x80f8c078, 0xb8ef1605,
+	0x806f816f, 0x8e6f846f,
+	0x8e76826f, 0xbef600ff,
+	0x01000000, 0xbefc006f,
+	0xc031003a, 0x00000078,
+	0x80f8c078, 0xbf8cc07f,
+	0x80fc907c, 0xbf800000,
+	0xbe802d00, 0xbe822d02,
+	0xbe842d04, 0xbe862d06,
+	0xbe882d08, 0xbe8a2d0a,
+	0xbe8c2d0c, 0xbe8e2d0e,
+	0xbf06807c, 0xbf84fff0,
+	0xb8f82985, 0x80788178,
+	0x8e788a78, 0x8e788178,
+	0xb8ee1605, 0x806e816e,
+	0x8e6e866e, 0x80786e78,
+	0xbef60084, 0xbef600ff,
+	0x01000000, 0xc0211bfa,
+	0x00000078, 0x80788478,
+	0xc0211b3a, 0x00000078,
+	0x80788478, 0xc0211b7a,
+	0x00000078, 0x80788478,
+	0xc0211c3a, 0x00000078,
+	0x80788478, 0xc0211c7a,
+	0x00000078, 0x80788478,
+	0xc0211eba, 0x00000078,
+	0x80788478, 0xc0211efa,
+	0x00000078, 0x80788478,
+	0xc0211a3a, 0x00000078,
+	0x80788478, 0xc0211a7a,
+	0x00000078, 0x80788478,
+	0xc0211cfa, 0x00000078,
+	0x80788478, 0xbf8cc07f,
+	0xbefc006f, 0xbefe0070,
+	0xbeff0071, 0x866f7bff,
+	0x000003ff, 0xb96f4803,
+	0x866f7bff, 0xfffff800,
+	0x8f6f8b6f, 0xb96fa2c3,
+	0xb973f801, 0xb8ee2985,
+	0x806e816e, 0x8e6e8a6e,
+	0x8e6e816e, 0xb8ef1605,
+	0x806f816f, 0x8e6f866f,
+	0x806e6f6e, 0x806e746e,
+	0x826f8075, 0x866fff6f,
+	0x0000ffff, 0xc00b1c37,
+	0x00000050, 0xc00b1d37,
+	0x00000060, 0xc0031e77,
+	0x00000074, 0xbf8cc07f,
+	0x8f6e8b79, 0x866eff6e,
+	0x001f8000, 0xb96ef807,
+	0x866dff6d, 0x0000ffff,
+	0x86fe7e7e, 0x86ea6a6a,
+	0x8f6e837a, 0xb96ee0c2,
+	0xbf800002, 0xb97a0002,
+	0xbf8a0000, 0xbe801f6c,
+	0xbf810000, 0x00000000,
+};
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm
@ -33,15 +33,20 @@
 * aldebaran:
 *   cpp -DASIC_FAMILY=CHIP_ALDEBARAN cwsr_trap_handler_gfx9.asm -P -o aldebaran.sp3
 *   sp3 aldebaran.sp3 -hex aldebaran.hex
+ *
+ * gc_9_4_3:
+ *   cpp -DASIC_FAMILY=GC_9_4_3 cwsr_trap_handler_gfx9.asm -P -o gc_9_4_3.sp3
+ *   sp3 gc_9_4_3.sp3 -hex gc_9_4_3.hex
 */

 #define CHIP_VEGAM 18
 #define CHIP_ARCTURUS 23
 #define CHIP_ALDEBARAN 25
+#define CHIP_GC_9_4_3 26

 var ACK_SQC_STORE		    =	1		    //workaround for suspected SQC store bug causing incorrect stores under concurrency
 var SAVE_AFTER_XNACK_ERROR	    =	1		    //workaround for TCP store failure after XNACK error when ALLOW_REPLAY=0, for debugger
-var SINGLE_STEP_MISSED_WORKAROUND   =	1		    //workaround for lost MODE.DEBUG_EN exception when SAVECTX raised
+var SINGLE_STEP_MISSED_WORKAROUND   =	(ASIC_FAMILY <= CHIP_ALDEBARAN)	//workaround for lost MODE.DEBUG_EN exception when SAVECTX raised

 /**************************************************************************/
 /*			variables					  */
@ -77,6 +82,10 @@ var SQ_WAVE_TRAPSTS_ADDR_WATCH_MASK =	0x80
 var SQ_WAVE_TRAPSTS_ADDR_WATCH_SHIFT =	7
 var SQ_WAVE_TRAPSTS_MEM_VIOL_MASK   =	0x100
 var SQ_WAVE_TRAPSTS_MEM_VIOL_SHIFT  =	8
+var SQ_WAVE_TRAPSTS_HOST_TRAP_MASK  =	0x400000
+var SQ_WAVE_TRAPSTS_WAVE_BEGIN_MASK =	0x800000
+var SQ_WAVE_TRAPSTS_WAVE_END_MASK   =	0x1000000
+var SQ_WAVE_TRAPSTS_TRAP_AFTER_INST_MASK =  0x2000000
 var SQ_WAVE_TRAPSTS_PRE_SAVECTX_MASK	=   0x3FF
 var SQ_WAVE_TRAPSTS_PRE_SAVECTX_SHIFT	=   0x0
 var SQ_WAVE_TRAPSTS_PRE_SAVECTX_SIZE	=   10
@ -95,10 +104,10 @@ var SQ_WAVE_IB_STS_RCNT_FIRST_REPLAY_MASK	= 0x1F8000

 var SQ_WAVE_MODE_DEBUG_EN_MASK		=   0x800

-var TTMP11_SAVE_RCNT_FIRST_REPLAY_SHIFT	=   26			// bits [31:26] unused by SPI debug data
-var TTMP11_SAVE_RCNT_FIRST_REPLAY_MASK	=   0xFC000000
-var TTMP11_DEBUG_TRAP_ENABLED_SHIFT	=   23
-var TTMP11_DEBUG_TRAP_ENABLED_MASK	=   0x800000
+var TTMP_SAVE_RCNT_FIRST_REPLAY_SHIFT	=   26			// bits [31:26] unused by SPI debug data
+var TTMP_SAVE_RCNT_FIRST_REPLAY_MASK	=   0xFC000000
+var TTMP_DEBUG_TRAP_ENABLED_SHIFT	=   23
+var TTMP_DEBUG_TRAP_ENABLED_MASK	=   0x800000

 /*	Save	    */
 var S_SAVE_BUF_RSRC_WORD1_STRIDE	=   0x00040000		//stride is 4 bytes
@ -129,6 +138,11 @@ var s_save_alloc_size	    =	s_save_trapsts		//conflict
 var s_save_m0		    =	ttmp5
 var s_save_ttmps_lo	    =	s_save_tmp		//no conflict
 var s_save_ttmps_hi	    =	s_save_trapsts		//no conflict
+#if ASIC_FAMILY >= CHIP_GC_9_4_3
+var s_save_ib_sts       =	ttmp13
+#else
+var s_save_ib_sts       =	ttmp11
+#endif

 /*	Restore	    */
 var S_RESTORE_BUF_RSRC_WORD1_STRIDE	    =	S_SAVE_BUF_RSRC_WORD1_STRIDE
@ -215,9 +229,15 @@ L_NOT_HALTED:
    // Any concurrent SAVECTX will be handled upon re-entry once halted.

    // Check non-maskable exceptions. memory_violation, illegal_instruction
-    // and xnack_error exceptions always cause the wave to enter the trap
-    // handler.
-    s_and_b32       ttmp2, s_save_trapsts, SQ_WAVE_TRAPSTS_MEM_VIOL_MASK|SQ_WAVE_TRAPSTS_ILLEGAL_INST_MASK
+    // and debugger (host trap, wave start/end, trap after instruction)
+    // exceptions always cause the wave to enter the trap handler.
+    s_and_b32       ttmp2, s_save_trapsts,      \
+        SQ_WAVE_TRAPSTS_MEM_VIOL_MASK         | \
+        SQ_WAVE_TRAPSTS_ILLEGAL_INST_MASK     | \
+        SQ_WAVE_TRAPSTS_HOST_TRAP_MASK        | \
+        SQ_WAVE_TRAPSTS_WAVE_BEGIN_MASK       | \
+        SQ_WAVE_TRAPSTS_WAVE_END_MASK         | \
+        SQ_WAVE_TRAPSTS_TRAP_AFTER_INST_MASK
    s_cbranch_scc1  L_FETCH_2ND_TRAP

    // Check for maskable exceptions in trapsts.excp and trapsts.excp_hi.
@ -265,9 +285,9 @@ L_FETCH_2ND_TRAP:

    s_load_dword    ttmp2, [ttmp14, ttmp15], 0x10 glc:1 // debug trap enabled flag
    s_waitcnt       lgkmcnt(0)
-    s_lshl_b32      ttmp2, ttmp2, TTMP11_DEBUG_TRAP_ENABLED_SHIFT
-    s_andn2_b32     ttmp11, ttmp11, TTMP11_DEBUG_TRAP_ENABLED_MASK
-    s_or_b32        ttmp11, ttmp11, ttmp2
+    s_lshl_b32      ttmp2, ttmp2, TTMP_DEBUG_TRAP_ENABLED_SHIFT
+    s_andn2_b32     s_save_ib_sts, s_save_ib_sts, TTMP_DEBUG_TRAP_ENABLED_MASK
+    s_or_b32        s_save_ib_sts, s_save_ib_sts, ttmp2

    s_load_dwordx2  [ttmp2, ttmp3], [ttmp14, ttmp15], 0x0 glc:1 // second-level TBA
    s_waitcnt       lgkmcnt(0)
@ -1058,17 +1078,17 @@ function set_status_without_spi_prio(status, tmp)
 end

 function save_and_clear_ib_sts(tmp)
-    // Save IB_STS.FIRST_REPLAY[15] and IB_STS.RCNT[20:16] into unused space ttmp11[31:26].
+    // Save IB_STS.FIRST_REPLAY[15] and IB_STS.RCNT[20:16] into unused space s_save_ib_sts[31:26].
    s_getreg_b32    tmp, hwreg(HW_REG_IB_STS)
    s_and_b32       tmp, tmp, SQ_WAVE_IB_STS_RCNT_FIRST_REPLAY_MASK
-    s_lshl_b32      tmp, tmp, (TTMP11_SAVE_RCNT_FIRST_REPLAY_SHIFT - SQ_WAVE_IB_STS_FIRST_REPLAY_SHIFT)
-    s_andn2_b32     ttmp11, ttmp11, TTMP11_SAVE_RCNT_FIRST_REPLAY_MASK
-    s_or_b32        ttmp11, ttmp11, tmp
+    s_lshl_b32      tmp, tmp, (TTMP_SAVE_RCNT_FIRST_REPLAY_SHIFT - SQ_WAVE_IB_STS_FIRST_REPLAY_SHIFT)
+    s_andn2_b32     s_save_ib_sts, s_save_ib_sts, TTMP_SAVE_RCNT_FIRST_REPLAY_MASK
+    s_or_b32        s_save_ib_sts, s_save_ib_sts, tmp
    s_setreg_imm32_b32 hwreg(HW_REG_IB_STS), 0x0
 end

 function restore_ib_sts(tmp)
-    s_lshr_b32      tmp, ttmp11, (TTMP11_SAVE_RCNT_FIRST_REPLAY_SHIFT - SQ_WAVE_IB_STS_FIRST_REPLAY_SHIFT)
+    s_lshr_b32      tmp, s_save_ib_sts, (TTMP_SAVE_RCNT_FIRST_REPLAY_SHIFT - SQ_WAVE_IB_STS_FIRST_REPLAY_SHIFT)
    s_and_b32       tmp, tmp, SQ_WAVE_IB_STS_RCNT_FIRST_REPLAY_MASK
    s_setreg_b32    hwreg(HW_REG_IB_STS), tmp
 end
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@ -1462,6 +1462,7 @@ int kfd_get_gpu_cache_info(struct kfd_dev *kdev, struct kfd_gpu_cache_info **pca
 			num_of_cache_types = ARRAY_SIZE(vega20_cache_info);
 			break;
 		case IP_VERSION(9, 4, 2):
+		case IP_VERSION(9, 4, 3):
 			*pcache_info = aldebaran_cache_info;
 			num_of_cache_types = ARRAY_SIZE(aldebaran_cache_info);
 			break;
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@ -315,6 +315,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 			break;
 		/* Aldebaran */
 		case IP_VERSION(9, 4, 2):
+		case IP_VERSION(9, 4, 3):
 			gfx_target_version = 90010;
 			f2g = &aldebaran_kfd2kgd;
 			break;
@ -450,6 +451,10 @@ static void kfd_cwsr_init(struct kfd_dev *kfd)
 			BUILD_BUG_ON(sizeof(cwsr_trap_aldebaran_hex) > PAGE_SIZE);
 			kfd->cwsr_isa = cwsr_trap_aldebaran_hex;
 			kfd->cwsr_isa_size = sizeof(cwsr_trap_aldebaran_hex);
+		} else if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3)) {
+			BUILD_BUG_ON(sizeof(cwsr_trap_gfx9_4_3_hex) > PAGE_SIZE);
+			kfd->cwsr_isa = cwsr_trap_gfx9_4_3_hex;
+			kfd->cwsr_isa_size = sizeof(cwsr_trap_gfx9_4_3_hex);
 		} else if (KFD_GC_VERSION(kfd) < IP_VERSION(10, 1, 1)) {
 			BUILD_BUG_ON(sizeof(cwsr_trap_gfx9_hex) > PAGE_SIZE);
 			kfd->cwsr_isa = cwsr_trap_gfx9_hex;
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@ -135,6 +135,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 {
 	uint64_t addr;
 	struct v9_mqd *m;
+	struct amdgpu_device *adev = (struct amdgpu_device *)mm->dev->adev;

 	m = (struct v9_mqd *) mqd_mem_obj->cpu_ptr;
 	addr = mqd_mem_obj->gpu_addr;
@ -167,6 +168,20 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 	if (q->format == KFD_QUEUE_FORMAT_AQL) {
 		m->cp_hqd_aql_control =
 			1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT;
+		if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+			/* On GC 9.4.3, DW 41 is re-purposed as
+			 * compute_tg_chunk_size.
+			 * TODO: review this setting when active CUs in the
+			 * partition play a role
+			 */
+			m->compute_static_thread_mgmt_se6 = 1;
+		}
+	} else {
+		/* PM4 queue */
+		if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+			m->compute_static_thread_mgmt_se6 = 0;
+			/* TODO: program pm4_target_xcc */
+		}
 	}

 	if (q->tba_addr) {
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
@ -238,7 +238,8 @@ int pm_init(struct packet_manager *pm, struct device_queue_manager *dqm)
 		pm->pmf = &kfd_vi_pm_funcs;
 		break;
 	default:
-		if (KFD_GC_VERSION(dqm->dev) == IP_VERSION(9, 4, 2))
+		if (KFD_GC_VERSION(dqm->dev) == IP_VERSION(9, 4, 2) ||
+		    KFD_GC_VERSION(dqm->dev) == IP_VERSION(9, 4, 3))
 			pm->pmf = &kfd_aldebaran_pm_funcs;
 		else if (KFD_GC_VERSION(dqm->dev) >= IP_VERSION(9, 0, 1))
 			pm->pmf = &kfd_v9_pm_funcs;
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@ -207,7 +207,8 @@ enum cache_policy {
 #define KFD_GC_VERSION(dev) ((dev)->adev->ip_versions[GC_HWIP][0])
 #define KFD_IS_SOC15(dev)   ((KFD_GC_VERSION(dev)) >= (IP_VERSION(9, 0, 1)))
 #define KFD_SUPPORT_XNACK_PER_PROCESS(dev)\
-		(KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2))
+	((KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2)) ||	\
+	 (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3)))

 struct kfd_event_interrupt_class {
 	bool (*interrupt_isr)(struct kfd_dev *dev,
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@ -820,15 +820,14 @@ static void dm_dmub_outbox1_low_irq(void *interrupt_params)
 					DRM_ERROR("Failed to allocate dmub_hpd_wrk");
 					return;
 				}
-				dmub_hpd_wrk->dmub_notify = kzalloc(sizeof(struct dmub_notification), GFP_ATOMIC);
+				dmub_hpd_wrk->dmub_notify = kmemdup(&notify, sizeof(struct dmub_notification),
+								    GFP_ATOMIC);
 				if (!dmub_hpd_wrk->dmub_notify) {
 					kfree(dmub_hpd_wrk);
 					DRM_ERROR("Failed to allocate dmub_hpd_wrk->dmub_notify");
 					return;
 				}
 				INIT_WORK(&dmub_hpd_wrk->handle_hpd_work, dm_handle_hpd_work);
-				if (dmub_hpd_wrk->dmub_notify)
-					memcpy(dmub_hpd_wrk->dmub_notify, &notify, sizeof(struct dmub_notification));
 				dmub_hpd_wrk->adev = adev;
 				if (notify.type == DMUB_NOTIFICATION_HPD) {
 					plink = adev->dm.dc->links[notify.link_index];
@ -2300,9 +2299,9 @@ static int dm_late_init(void *handle)
 	 */
 	params.min_abm_backlight = 0x28F;
 	/* In the case where abm is implemented on dmcub,
-	* dmcu object will be null.
-	* ABM 2.4 and up are implemented on dmcub.
-	*/
+	 * dmcu object will be null.
+	 * ABM 2.4 and up are implemented on dmcub.
+	 */
 	if (dmcu) {
 		if (!dmcu_load_iram(dmcu, params))
 			return -EINVAL;
@ -2976,30 +2975,18 @@ static struct drm_mode_config_helper_funcs amdgpu_dm_mode_config_helperfuncs = {
 static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 {
 	struct amdgpu_dm_backlight_caps *caps;
-	struct amdgpu_display_manager *dm;
 	struct drm_connector *conn_base;
 	struct amdgpu_device *adev;
-	struct dc_link *link = NULL;
 	struct drm_luminance_range_info *luminance_range;
-	int i;

-	if (!aconnector || !aconnector->dc_link)
-		return;
-
-	link = aconnector->dc_link;
-	if (link->connector_signal != SIGNAL_TYPE_EDP)
+	if (aconnector->bl_idx == -1 ||
+	    aconnector->dc_link->connector_signal != SIGNAL_TYPE_EDP)
 		return;

 	conn_base = &aconnector->base;
 	adev = drm_to_adev(conn_base->dev);
-	dm = &adev->dm;
-	for (i = 0; i < dm->num_of_edps; i++) {
-		if (link == dm->backlight_link[i])
-			break;
-	}
-	if (i >= dm->num_of_edps)
-		return;
-	caps = &dm->backlight_caps[i];
+
+	caps = &adev->dm.backlight_caps[aconnector->bl_idx];
 	caps->ext_caps = &aconnector->dc_link->dpcd_sink_ext_caps;
 	caps->aux_support = false;

@ -4190,16 +4177,18 @@ static const struct backlight_ops amdgpu_dm_backlight_ops = {
 };

 static void
-amdgpu_dm_register_backlight_device(struct amdgpu_display_manager *dm)
+amdgpu_dm_register_backlight_device(struct amdgpu_dm_connector *aconnector)
 {
-	char bl_name[16];
+	struct drm_device *drm = aconnector->base.dev;
+	struct amdgpu_display_manager *dm = &drm_to_adev(drm)->dm;
 	struct backlight_properties props = { 0 };
+	char bl_name[16];

-	amdgpu_dm_update_backlight_caps(dm, dm->num_of_edps);
-	dm->brightness[dm->num_of_edps] = AMDGPU_MAX_BL_LEVEL;
+	if (aconnector->bl_idx == -1)
+		return;

 	if (!acpi_video_backlight_use_native()) {
-		drm_info(adev_to_drm(dm->adev), "Skipping amdgpu DM backlight registration\n");
+		drm_info(drm, "Skipping amdgpu DM backlight registration\n");
 		/* Try registering an ACPI video backlight device instead. */
 		acpi_video_register_backlight();
 		return;
@ -4210,17 +4199,16 @@ amdgpu_dm_register_backlight_device(struct amdgpu_display_manager *dm)
 	props.type = BACKLIGHT_RAW;

 	snprintf(bl_name, sizeof(bl_name), "amdgpu_bl%d",
-		 adev_to_drm(dm->adev)->primary->index + dm->num_of_edps);
+		 drm->primary->index + aconnector->bl_idx);

-	dm->backlight_dev[dm->num_of_edps] = backlight_device_register(bl_name,
-								       adev_to_drm(dm->adev)->dev,
-								       dm,
-								       &amdgpu_dm_backlight_ops,
-								       &props);
+	dm->backlight_dev[aconnector->bl_idx] =
+		backlight_device_register(bl_name, aconnector->base.kdev, dm,
+					  &amdgpu_dm_backlight_ops, &props);

-	if (IS_ERR(dm->backlight_dev[dm->num_of_edps]))
+	if (IS_ERR(dm->backlight_dev[aconnector->bl_idx])) {
 		DRM_ERROR("DM: Backlight registration failed!\n");
-	else
+		dm->backlight_dev[aconnector->bl_idx] = NULL;
+	} else
 		DRM_DEBUG_DRIVER("DM: Registered Backlight device: %s\n", bl_name);
 }

@ -4265,24 +4253,29 @@ static int initialize_plane(struct amdgpu_display_manager *dm,
 }


-static void register_backlight_device(struct amdgpu_display_manager *dm,
-				      struct dc_link *link)
+static void setup_backlight_device(struct amdgpu_display_manager *dm,
+				   struct amdgpu_dm_connector *aconnector)
 {
-	if ((link->connector_signal & (SIGNAL_TYPE_EDP | SIGNAL_TYPE_LVDS)) &&
-	    link->type != dc_connection_none) {
-		/*
-		 * Event if registration failed, we should continue with
-		 * DM initialization because not having a backlight control
-		 * is better then a black screen.
-		 */
-		if (!dm->backlight_dev[dm->num_of_edps])
-			amdgpu_dm_register_backlight_device(dm);
+	struct dc_link *link = aconnector->dc_link;
+	int bl_idx = dm->num_of_edps;

-		if (dm->backlight_dev[dm->num_of_edps]) {
-			dm->backlight_link[dm->num_of_edps] = link;
-			dm->num_of_edps++;
-		}
+	if (!(link->connector_signal & (SIGNAL_TYPE_EDP | SIGNAL_TYPE_LVDS)) ||
+	    link->type == dc_connection_none)
+		return;
+
+	if (dm->num_of_edps >= AMDGPU_DM_MAX_NUM_EDP) {
+		drm_warn(adev_to_drm(dm->adev), "Too much eDP connections, skipping backlight setup for additional eDPs\n");
+		return;
 	}
+
+	aconnector->bl_idx = bl_idx;
+
+	amdgpu_dm_update_backlight_caps(dm, bl_idx);
+	dm->brightness[bl_idx] = AMDGPU_MAX_BL_LEVEL;
+	dm->backlight_link[bl_idx] = link;
+	dm->num_of_edps++;
+
+	update_connector_ext_caps(aconnector);
 }

 static void amdgpu_set_panel_orientation(struct drm_connector *connector);
@ -4462,10 +4455,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)

 			if (ret) {
 				amdgpu_dm_update_connector_after_detect(aconnector);
-				register_backlight_device(dm, link);
-
-				if (dm->num_of_edps)
-					update_connector_ext_caps(aconnector);
+				setup_backlight_device(dm, aconnector);

 				if (psr_feature_enabled)
 					amdgpu_dm_set_psr_caps(link);
@ -6241,10 +6231,8 @@ static void amdgpu_dm_connector_unregister(struct drm_connector *connector)
 static void amdgpu_dm_connector_destroy(struct drm_connector *connector)
 {
 	struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
-	const struct dc_link *link = aconnector->dc_link;
 	struct amdgpu_device *adev = drm_to_adev(connector->dev);
 	struct amdgpu_display_manager *dm = &adev->dm;
-	int i;

 	/*
 	 * Call only if mst_mgr was initialized before since it's not done
@ -6253,11 +6241,9 @@ static void amdgpu_dm_connector_destroy(struct drm_connector *connector)
 	if (aconnector->mst_mgr.dev)
 		drm_dp_mst_topology_mgr_destroy(&aconnector->mst_mgr);

-	for (i = 0; i < dm->num_of_edps; i++) {
-		if ((link == dm->backlight_link[i]) && dm->backlight_dev[i]) {
-			backlight_device_unregister(dm->backlight_dev[i]);
-			dm->backlight_dev[i] = NULL;
-		}
+	if (aconnector->bl_idx != -1) {
+		backlight_device_unregister(dm->backlight_dev[aconnector->bl_idx]);
+		dm->backlight_dev[aconnector->bl_idx] = NULL;
 	}

 	if (aconnector->dc_em_sink)
@ -6339,6 +6325,8 @@ amdgpu_dm_connector_late_register(struct drm_connector *connector)
 		to_amdgpu_dm_connector(connector);
 	int r;

+	amdgpu_dm_register_backlight_device(amdgpu_dm_connector);
+
 	if ((connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) ||
 	    (connector->connector_type == DRM_MODE_CONNECTOR_eDP)) {
 		amdgpu_dm_connector->dm_dp_aux.aux.dev = connector->kdev;
@ -7103,13 +7091,13 @@ static uint add_fs_modes(struct amdgpu_dm_connector *aconnector)
 	/* Standard FPS values
 	 *
 	 * 23.976       - TV/NTSC
-	 * 24 	        - Cinema
-	 * 25 	        - TV/PAL
+	 * 24           - Cinema
+	 * 25           - TV/PAL
 	 * 29.97        - TV/NTSC
-	 * 30 	        - TV/NTSC
-	 * 48 	        - Cinema HFR
-	 * 50 	        - TV/PAL
-	 * 60 	        - Commonly used
+	 * 30           - TV/NTSC
+	 * 48           - Cinema HFR
+	 * 50           - TV/PAL
+	 * 60           - Commonly used
 	 * 48,72,96,120 - Multiples of 24
 	 */
 	static const u32 common_rates[] = {
@ -7227,6 +7215,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
 		aconnector->base.funcs->reset(&aconnector->base);

 	aconnector->connector_id = link_index;
+	aconnector->bl_idx = -1;
 	aconnector->dc_link = link;
 	aconnector->base.interlace_allowed = false;
 	aconnector->base.doublescan_allowed = false;
@ -7737,7 +7726,7 @@ static void update_freesync_state_on_stream(
 		return;

 	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
-        vrr_params = acrtc->dm_irq_params.vrr_params;
+	vrr_params = acrtc->dm_irq_params.vrr_params;

 	if (surface) {
 		mod_freesync_handle_preflip(
@ -8318,7 +8307,7 @@ static void amdgpu_dm_commit_audio(struct drm_device *dev,
 		if (!drm_atomic_crtc_needs_modeset(new_crtc_state))
 			continue;

-	notify:
+notify:
 		aconnector = to_amdgpu_dm_connector(connector);

 		mutex_lock(&adev->dm.audio_lock);
@ -9334,7 +9323,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 skip_modeset:
 	/* Release extra reference */
 	if (new_stream)
-		 dc_stream_release(new_stream);
+		dc_stream_release(new_stream);

 	/*
 	 * We want to do dc stream updates that do not require a
@ -10668,7 +10657,7 @@ int amdgpu_dm_process_dmub_aux_transfer_sync(
 	if (!dc_process_dmub_aux_transfer_async(ctx->dc, link_index, payload)) {
 		*operation_result = AUX_RET_ERROR_ENGINE_ACQUIRE;
 		goto out;
- 	}
+	}

 	if (!wait_for_completion_timeout(&adev->dm.dmub_aux_transfer_done, 10 * HZ)) {
 		DRM_ERROR("wait_for_completion_timeout timeout!");
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@ -610,6 +610,7 @@ struct amdgpu_dm_connector {

 	struct drm_connector base;
 	uint32_t connector_id;
+	int bl_idx;

 	/* we need to mind the EDID between detect
 	   and get modes due to analog/digital/tvencoder */
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@ -44,6 +44,9 @@
 #include "dm_helpers.h"
 #include "ddc_service_types.h"

+/* MST Dock */
+static const uint8_t SYNAPTICS_DEVICE_ID[] = "SYNA";
+
 /* dm_helpers_parse_edid_caps
 *
 * Parse edid caps
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@ -208,7 +208,7 @@ bool needs_dsc_aux_workaround(struct dc_link *link)
 	return false;
 }

-bool is_synaptics_cascaded_panamera(struct dc_link *link, struct drm_dp_mst_port *port)
+static bool is_synaptics_cascaded_panamera(struct dc_link *link, struct drm_dp_mst_port *port)
 {
 	u8 branch_vendor_data[4] = { 0 }; // Vendor data 0x50C ~ 0x50F

--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
@ -522,6 +522,11 @@ static void dcn315_clk_mgr_helper_populate_bw_params(
 		bw_params->clk_table.entries[i].dcfclk_mhz = clock_table->DcfClocks[0];
 		bw_params->clk_table.entries[i].wck_ratio = 1;
 		i++;
+	} else if (clock_table->NumDcfClkLevelsEnabled != clock_table->NumSocClkLevelsEnabled) {
+		bw_params->clk_table.entries[i-1].voltage = clock_table->SocVoltage[clock_table->NumSocClkLevelsEnabled - 1];
+		bw_params->clk_table.entries[i-1].socclk_mhz = clock_table->SocClocks[clock_table->NumSocClkLevelsEnabled - 1];
+		bw_params->clk_table.entries[i-1].dispclk_mhz = clock_table->DispClocks[clock_table->NumDispClkLevelsEnabled - 1];
+		bw_params->clk_table.entries[i-1].dppclk_mhz = clock_table->DppClocks[clock_table->NumDispClkLevelsEnabled - 1];
 	}
 	bw_params->clk_table.num_entries = i;

--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@ -888,6 +888,10 @@ static bool dc_construct_ctx(struct dc *dc,

 	dc->ctx = dc_ctx;

+	dc->link_srv = link_create_link_service();
+	if (!dc->link_srv)
+		return false;
+
 	return true;
 }

@ -985,8 +989,6 @@ static bool dc_construct(struct dc *dc,
 		goto fail;
 	}

-	dc->link_srv = link_create_link_service();
-
 	dc->res_pool = dc_create_resource_pool(dc, init_params, dc_ctx->dce_version);
 	if (!dc->res_pool)
 		goto fail;
@ -3492,22 +3494,6 @@ static void commit_planes_for_stream(struct dc *dc,

 	dc_dmub_update_dirty_rect(dc, surface_count, stream, srf_updates, context);

-	if (update_type != UPDATE_TYPE_FAST) {
-		for (i = 0; i < dc->res_pool->pipe_count; i++) {
-			struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i];
-
-			if ((new_pipe->stream && new_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) ||
-					subvp_prev_use) {
-				// If old context or new context has phantom pipes, apply
-				// the phantom timings now. We can't change the phantom
-				// pipe configuration safely without driver acquiring
-				// the DMCUB lock first.
-				dc->hwss.apply_ctx_to_hw(dc, context);
-				break;
-			}
-		}
-	}
-
 	// Stream updates
 	if (stream_update)
 		commit_planes_do_stream_update(dc, stream, stream_update, update_type, context);
@ -3723,6 +3709,9 @@ static void commit_planes_for_stream(struct dc *dc,
 		}
 	}

+	if (update_type != UPDATE_TYPE_FAST)
+		dc->hwss.post_unlock_program_front_end(dc, context);
+
 	if (subvp_prev_use && !subvp_curr_use) {
 		/* If disabling subvp, disable phantom streams after front end
 		 * programming has completed (we turn on phantom OTG in order
@ -3731,16 +3720,9 @@ static void commit_planes_for_stream(struct dc *dc,
 		dc->hwss.apply_ctx_to_hw(dc, context);
 	}

-	if (update_type != UPDATE_TYPE_FAST)
-		dc->hwss.post_unlock_program_front_end(dc, context);
 	if (update_type != UPDATE_TYPE_FAST)
 		if (dc->hwss.commit_subvp_config)
 			dc->hwss.commit_subvp_config(dc, context);
-
-	if (update_type != UPDATE_TYPE_FAST)
-		if (dc->hwss.commit_subvp_config)
-			dc->hwss.commit_subvp_config(dc, context);
-
 	/* Since phantom pipe programming is moved to post_unlock_program_front_end,
 	 * move the SubVP lock to after the phantom pipes have been setup
 	 */
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
@ -285,8 +285,7 @@ int dc_link_aux_transfer_raw(struct ddc_service *ddc,
 			ddc, payload, operation_result);
 }

-uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(
-		struct dc *dc, uint8_t bw)
+uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(const struct dc *dc, uint8_t bw)
 {
 	return dc->link_srv->bw_kbps_from_raw_frl_link_rate_data(bw);
 }
@ -474,3 +473,8 @@ void dc_link_enable_hpd_filter(struct dc_link *link, bool enable)
 {
 	link->dc->link_srv->enable_hpd_filter(link, enable);
 }
+
+bool dc_link_validate(struct dc *dc, const struct dc_stream_state *streams, const unsigned int count)
+{
+	return dc->link_srv->validate_dpia_bandwidth(streams, count);
+}
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stat.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stat.c
@ -35,19 +35,15 @@
 */

 /**
- *****************************************************************************
- *  Function: dc_stat_get_dmub_notification
+ *  dc_stat_get_dmub_notification
 *
- *  @brief
- *		Calls dmub layer to retrieve dmub notification
+ * Calls dmub layer to retrieve dmub notification
 *
- *  @param
- *		[in] dc: dc structure
- *		[in] notify: dmub notification structure
+ * @dc: dc structure
+ * @notify: dmub notification structure
 *
- *  @return
+ * Returns
 *     None
- *****************************************************************************
 */
 void dc_stat_get_dmub_notification(const struct dc *dc, struct dmub_notification *notify)
 {
@ -73,19 +69,15 @@ void dc_stat_get_dmub_notification(const struct dc *dc, struct dmub_notification
 }

 /**
- *****************************************************************************
- *  Function: dc_stat_get_dmub_dataout
+ * dc_stat_get_dmub_dataout
 *
- *  @brief
- *		Calls dmub layer to retrieve dmub gpint dataout
+ * Calls dmub layer to retrieve dmub gpint dataout
 *
- *  @param
- *		[in] dc: dc structure
- *		[in] dataout: dmub gpint dataout
+ * @dc: dc structure
+ * @dataout: dmub gpint dataout
 *
- *  @return
+ * Returns
 *     None
- *****************************************************************************
 */
 void dc_stat_get_dmub_dataout(const struct dc *dc, uint32_t *dataout)
 {
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;

-#define DC_VER "3.2.227"
+#define DC_VER "3.2.229"

 #define MAX_SURFACES 3
 #define MAX_PLANES 6
@ -873,6 +873,8 @@ struct dc_debug_options {
 	bool dig_fifo_off_in_blank;
 	bool temp_mst_deallocation_sequence;
 	bool override_dispclk_programming;
+	bool disable_fpo_optimizations;
+	bool support_eDP1_5;
 };

 struct gpu_info_soc_bounding_box_v1_0;
@ -1768,8 +1770,7 @@ void dc_restore_link_res_map(const struct dc *dc, uint32_t *map);
 bool dc_link_update_dsc_config(struct pipe_ctx *pipe_ctx);

 /* translate a raw link rate data to bandwidth in kbps */
-uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(
-		struct dc *dc, uint8_t bw);
+uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(const struct dc *dc, uint8_t bw);

 /* determine the optimal bandwidth given link and required bw.
 * @link - current detected link
@ -2029,6 +2030,19 @@ void dc_link_handle_usb4_bw_alloc_response(struct dc_link *link,
 int dc_link_dp_dpia_handle_usb4_bandwidth_allocation_for_link(
 		struct dc_link *link, int peak_bw);

+/*
+ * Validate the BW of all the valid DPIA links to make sure it doesn't exceed
+ * available BW for each host router
+ *
+ * @dc: pointer to dc struct
+ * @stream: pointer to all possible streams
+ * @num_streams: number of valid DPIA streams
+ *
+ * return: TRUE if bw used by DPIAs doesn't exceed available BW else return FALSE
+ */
+bool dc_link_validate(struct dc *dc, const struct dc_stream_state *streams,
+		const unsigned int count);
+
 /* Sink Interfaces - A sink corresponds to a display output device */

 struct dc_container_id {
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@ -302,27 +302,29 @@ static uint8_t dc_dmub_srv_get_pipes_for_stream(struct dc *dc, struct dc_stream_
 	return pipes;
 }

-static int dc_dmub_srv_get_timing_generator_offset(struct dc *dc, struct dc_stream_state *stream)
+static void dc_dmub_srv_populate_fams_pipe_info(struct dc *dc, struct dc_state *context,
+		struct pipe_ctx *head_pipe,
+		struct dmub_cmd_fw_assisted_mclk_switch_pipe_data *fams_pipe_data)
 {
-	int  tg_inst = 0;
-	int i = 0;
+	int j;
+	int pipe_idx = 0;

-	for (i = 0; i < MAX_PIPES; i++) {
-		struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+	fams_pipe_data->pipe_index[pipe_idx++] = head_pipe->plane_res.hubp->inst;
+	for (j = 0; j < dc->res_pool->pipe_count; j++) {
+		struct pipe_ctx *split_pipe = &context->res_ctx.pipe_ctx[j];

-		if (pipe->stream == stream && pipe->stream_res.tg) {
-			tg_inst = pipe->stream_res.tg->inst;
-			break;
+		if (split_pipe->stream == head_pipe->stream && (split_pipe->top_pipe || split_pipe->prev_odm_pipe)) {
+			fams_pipe_data->pipe_index[pipe_idx++] = split_pipe->plane_res.hubp->inst;
 		}
 	}
-	return tg_inst;
+	fams_pipe_data->pipe_count = pipe_idx;
 }

 bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, struct dc_state *context)
 {
 	union dmub_rb_cmd cmd = { 0 };
 	struct dmub_cmd_fw_assisted_mclk_switch_config *config_data = &cmd.fw_assisted_mclk_switch.config_data;
-	int i = 0;
+	int i = 0, k = 0;
 	int ramp_up_num_steps = 1; // TODO: Ramp is currently disabled. Reenable it.
 	uint8_t visual_confirm_enabled;

@ -337,17 +339,21 @@ bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, stru
 	cmd.fw_assisted_mclk_switch.config_data.fams_enabled = should_manage_pstate;
 	cmd.fw_assisted_mclk_switch.config_data.visual_confirm_enabled = visual_confirm_enabled;

-	for (i = 0; context && i < context->stream_count; i++) {
-		struct dc_stream_state *stream = context->streams[i];
-		uint8_t min_refresh_in_hz = (stream->timing.min_refresh_in_uhz + 999999) / 1000000;
-		int  tg_inst = dc_dmub_srv_get_timing_generator_offset(dc, stream);
+	for (i = 0, k = 0; context && i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];

-		config_data->pipe_data[tg_inst].pix_clk_100hz = stream->timing.pix_clk_100hz;
-		config_data->pipe_data[tg_inst].min_refresh_in_hz = min_refresh_in_hz;
-		config_data->pipe_data[tg_inst].max_ramp_step = ramp_up_num_steps;
-		config_data->pipe_data[tg_inst].pipes = dc_dmub_srv_get_pipes_for_stream(dc, stream);
+		if (!pipe->top_pipe && !pipe->prev_odm_pipe && pipe->stream && pipe->stream->fpo_in_use) {
+			struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+			uint8_t min_refresh_in_hz = (pipe->stream->timing.min_refresh_in_uhz + 999999) / 1000000;
+
+			config_data->pipe_data[k].pix_clk_100hz = pipe->stream->timing.pix_clk_100hz;
+			config_data->pipe_data[k].min_refresh_in_hz = min_refresh_in_hz;
+			config_data->pipe_data[k].max_ramp_step = ramp_up_num_steps;
+			config_data->pipe_data[k].pipes = dc_dmub_srv_get_pipes_for_stream(dc, pipe->stream);
+			dc_dmub_srv_populate_fams_pipe_info(dc, context, pipe, &config_data->pipe_data[k]);
+			k++;
+		}
 	}
-
 	cmd.fw_assisted_mclk_switch.header.payload_bytes =
 		sizeof(cmd.fw_assisted_mclk_switch) - sizeof(cmd.fw_assisted_mclk_switch.header);

--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@ -47,14 +47,15 @@ enum dc_lane_count {
 */
 enum dc_link_rate {
 	LINK_RATE_UNKNOWN = 0,
-	LINK_RATE_LOW = 0x06,		// Rate_1 (RBR)	- 1.62 Gbps/Lane
-	LINK_RATE_RATE_2 = 0x08,	// Rate_2		- 2.16 Gbps/Lane
-	LINK_RATE_RATE_3 = 0x09,	// Rate_3		- 2.43 Gbps/Lane
-	LINK_RATE_HIGH = 0x0A,		// Rate_4 (HBR)	- 2.70 Gbps/Lane
-	LINK_RATE_RBR2 = 0x0C,		// Rate_5 (RBR2)- 3.24 Gbps/Lane
-	LINK_RATE_RATE_6 = 0x10,	// Rate_6		- 4.32 Gbps/Lane
-	LINK_RATE_HIGH2 = 0x14,		// Rate_7 (HBR2)- 5.40 Gbps/Lane
-	LINK_RATE_HIGH3 = 0x1E,		// Rate_8 (HBR3)- 8.10 Gbps/Lane
+	LINK_RATE_LOW = 0x06,		// Rate_1 (RBR)  - 1.62 Gbps/Lane
+	LINK_RATE_RATE_2 = 0x08,	// Rate_2        - 2.16 Gbps/Lane
+	LINK_RATE_RATE_3 = 0x09,	// Rate_3        - 2.43 Gbps/Lane
+	LINK_RATE_HIGH = 0x0A,		// Rate_4 (HBR)  - 2.70 Gbps/Lane
+	LINK_RATE_RBR2 = 0x0C,		// Rate_5 (RBR2) - 3.24 Gbps/Lane
+	LINK_RATE_RATE_6 = 0x10,	// Rate_6        - 4.32 Gbps/Lane
+	LINK_RATE_HIGH2 = 0x14,		// Rate_7 (HBR2) - 5.40 Gbps/Lane
+	LINK_RATE_RATE_8 = 0x19,	// Rate_8        - 6.75 Gbps/Lane
+	LINK_RATE_HIGH3 = 0x1E,		// Rate_9 (HBR3) - 8.10 Gbps/Lane
 	/* Starting from DP2.0 link rate enum directly represents actual
 	 * link rate value in unit of 10 mbps
 	 */
--- a/drivers/gpu/drm/amd/display/dc/dc_hdmi_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_hdmi_types.h
@ -50,7 +50,6 @@ struct dp_hdmi_dongle_signature_data {

 /* DP-HDMI dongle slave address for retrieving dongle signature*/
 #define DP_HDMI_DONGLE_ADDRESS 0x40
-static const uint8_t dp_hdmi_dongle_signature_str[] = "DP-HDMI ADAPTOR";
 #define DP_HDMI_DONGLE_SIGNATURE_EOT 0x04


--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@ -293,6 +293,7 @@ struct dc_stream_state {

 	bool has_non_synchronizable_pclk;
 	bool vblank_synchronized;
+	bool fpo_in_use;
 	struct mall_stream_config mall_stream_config;
 };

--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
@ -1157,6 +1157,7 @@ const struct pixel_rate_range_table_entry video_optimized_pixel_rates[] = {
 	{25170, 25180, 25200, 1000, 1001},	//25.2MHz   ->   25.17
 	{59340, 59350, 59400, 1000, 1001},	//59.4Mhz   ->   59.340
 	{74170, 74180, 74250, 1000, 1001},	//74.25Mhz  ->   74.1758
+	{89910, 90000, 90000, 1000, 1001},	//90Mhz     ->   89.91
 	{125870, 125880, 126000, 1000, 1001},	//126Mhz    ->  125.87
 	{148350, 148360, 148500, 1000, 1001},	//148.5Mhz  ->  148.3516
 	{167830, 167840, 168000, 1000, 1001},	//168Mhz    ->  167.83
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
@ -33,6 +33,9 @@

 #define MAX_PIPES 6

+static const uint8_t DP_SINK_DEVICE_STR_ID_1[] = {7, 1, 8, 7, 3};
+static const uint8_t DP_SINK_DEVICE_STR_ID_2[] = {7, 1, 8, 7, 5};
+
 /*
 * Convert dmcub psr state to dmcu psr state.
 */
@ -250,7 +253,7 @@ static void dmub_psr_set_level(struct dmub_psr *dmub, uint16_t psr_level, uint8_
 	dc_dmub_srv_wait_idle(dc->dmub_srv);
 }

-/**
+/*
 * Set PSR vtotal requirement for FreeSync PSR.
 */
 static void dmub_psr_set_sink_vtotal_in_psr_active(struct dmub_psr *dmub,
--- a/drivers/gpu/drm/amd/display/dc/dce60/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dce60/Makefile
@ -23,7 +23,7 @@
 # Makefile for the 'controller' sub-component of DAL.
 # It provides the control and status of HW CRTC block.

-CFLAGS_AMDDALPATH)/dc/dce60/dce60_resource.o = $(call cc-disable-warning, override-init)
+CFLAGS_$(AMDDALPATH)/dc/dce60/dce60_resource.o = $(call cc-disable-warning, override-init)

 DCE60 = dce60_timing_generator.o dce60_hw_sequencer.o \
 	dce60_resource.o
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@ -726,11 +726,15 @@ void dcn10_hubp_pg_control(
 	}
 }

-static void power_on_plane(
+static void power_on_plane_resources(
 	struct dce_hwseq *hws,
 	int plane_id)
 {
 	DC_LOGGER_INIT(hws->ctx->logger);
+
+	if (hws->funcs.dpp_root_clock_control)
+		hws->funcs.dpp_root_clock_control(hws, plane_id, true);
+
 	if (REG(DC_IP_REQUEST_CNTL)) {
 		REG_SET(DC_IP_REQUEST_CNTL, 0,
 				IP_REQUEST_EN, 1);
@ -1237,11 +1241,15 @@ void dcn10_plane_atomic_power_down(struct dc *dc,
 			hws->funcs.hubp_pg_control(hws, hubp->inst, false);

 		dpp->funcs->dpp_reset(dpp);
+
 		REG_SET(DC_IP_REQUEST_CNTL, 0,
 				IP_REQUEST_EN, 0);
 		DC_LOG_DEBUG(
 				"Power gated front end %d\n", hubp->inst);
 	}
+
+	if (hws->funcs.dpp_root_clock_control)
+		hws->funcs.dpp_root_clock_control(hws, dpp->inst, false);
 }

 /* disable HW used by plane.
@ -2462,7 +2470,7 @@ static void dcn10_enable_plane(

 	undo_DEGVIDCN10_253_wa(dc);

-	power_on_plane(dc->hwseq,
+	power_on_plane_resources(dc->hwseq,
 		pipe_ctx->plane_res.hubp->inst);

 	/* enable DCFCLK current DCHUB */
@ -3385,7 +3393,9 @@ static bool dcn10_can_pipe_disable_cursor(struct pipe_ctx *pipe_ctx)
 	for (test_pipe = pipe_ctx->top_pipe; test_pipe;
 	     test_pipe = test_pipe->top_pipe) {
 		// Skip invisible layer and pipe-split plane on same layer
-		if (!test_pipe->plane_state->visible || test_pipe->plane_state->layer_index == cur_layer)
+		if (!test_pipe->plane_state ||
+		    !test_pipe->plane_state->visible ||
+		    test_pipe->plane_state->layer_index == cur_layer)
 			continue;

 		r2 = test_pipe->plane_res.scl_data.recout;
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
@ -161,10 +161,20 @@ struct dcn_optc_registers {
 	uint32_t OTG_CRC_CNTL2;
 	uint32_t OTG_CRC0_DATA_RG;
 	uint32_t OTG_CRC0_DATA_B;
+	uint32_t OTG_CRC1_DATA_B;
+	uint32_t OTG_CRC2_DATA_B;
+	uint32_t OTG_CRC3_DATA_B;
+	uint32_t OTG_CRC1_DATA_RG;
+	uint32_t OTG_CRC2_DATA_RG;
+	uint32_t OTG_CRC3_DATA_RG;
 	uint32_t OTG_CRC0_WINDOWA_X_CONTROL;
 	uint32_t OTG_CRC0_WINDOWA_Y_CONTROL;
 	uint32_t OTG_CRC0_WINDOWB_X_CONTROL;
 	uint32_t OTG_CRC0_WINDOWB_Y_CONTROL;
+	uint32_t OTG_CRC1_WINDOWA_X_CONTROL;
+	uint32_t OTG_CRC1_WINDOWA_Y_CONTROL;
+	uint32_t OTG_CRC1_WINDOWB_X_CONTROL;
+	uint32_t OTG_CRC1_WINDOWB_Y_CONTROL;
 	uint32_t GSL_SOURCE_SELECT;
 	uint32_t DWB_SOURCE_SELECT;
 	uint32_t OTG_DSC_START_POSITION;
@ -464,6 +474,15 @@ struct dcn_optc_registers {
 	type CRC0_R_CR;\
 	type CRC0_G_Y;\
 	type CRC0_B_CB;\
+	type CRC1_R_CR;\
+	type CRC1_G_Y;\
+	type CRC1_B_CB;\
+	type CRC2_R_CR;\
+	type CRC2_G_Y;\
+	type CRC2_B_CB;\
+	type CRC3_R_CR;\
+	type CRC3_G_Y;\
+	type CRC3_B_CB;\
 	type OTG_CRC0_WINDOWA_X_START;\
 	type OTG_CRC0_WINDOWA_X_END;\
 	type OTG_CRC0_WINDOWA_Y_START;\
@ -472,6 +491,15 @@ struct dcn_optc_registers {
 	type OTG_CRC0_WINDOWB_X_END;\
 	type OTG_CRC0_WINDOWB_Y_START;\
 	type OTG_CRC0_WINDOWB_Y_END;\
+	type OTG_CRC_WINDOW_DB_EN;\
+	type OTG_CRC1_WINDOWA_X_START;\
+	type OTG_CRC1_WINDOWA_X_END;\
+	type OTG_CRC1_WINDOWA_Y_START;\
+	type OTG_CRC1_WINDOWA_Y_END;\
+	type OTG_CRC1_WINDOWB_X_START;\
+	type OTG_CRC1_WINDOWB_X_END;\
+	type OTG_CRC1_WINDOWB_Y_START;\
+	type OTG_CRC1_WINDOWB_Y_END;\
 	type GSL0_READY_SOURCE_SEL;\
 	type GSL1_READY_SOURCE_SEL;\
 	type GSL2_READY_SOURCE_SEL;\
@ -525,6 +553,7 @@ struct dcn_optc_registers {
 #define TG_REG_FIELD_LIST_DCN3_2(type) \
 	type OTG_H_TIMING_DIV_MODE_MANUAL;

+
 struct dcn_optc_shift {
 	TG_REG_FIELD_LIST(uint8_t)
 	TG_REG_FIELD_LIST_DCN3_2(uint8_t)
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@ -542,8 +542,8 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.disable_pplib_clock_request = false,
 		.disable_pplib_wm_range = false,
 		.pplib_wm_report_mode = WM_REPORT_DEFAULT,
-		.pipe_split_policy = MPC_SPLIT_AVOID,
-		.force_single_disp_pipe_split = false,
+		.pipe_split_policy = MPC_SPLIT_DYNAMIC,
+		.force_single_disp_pipe_split = true,
 		.disable_dcc = DCC_ENABLE,
 		.voltage_align_fclk = true,
 		.disable_stereo_support = true,
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
@ -230,7 +230,8 @@
 	type DTBCLK_P2_SRC_SEL;\
 	type DTBCLK_P2_EN;\
 	type DTBCLK_P3_SRC_SEL;\
-	type DTBCLK_P3_EN;
+	type DTBCLK_P3_EN;\
+	type DENTIST_DISPCLK_CHG_DONE;

 struct dccg_shift {
 	DCCG_REG_FIELD_LIST(uint8_t)
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@ -1130,11 +1130,15 @@ void dcn20_blank_pixel_data(
 }


-static void dcn20_power_on_plane(
+static void dcn20_power_on_plane_resources(
 	struct dce_hwseq *hws,
 	struct pipe_ctx *pipe_ctx)
 {
 	DC_LOGGER_INIT(hws->ctx->logger);
+
+	if (hws->funcs.dpp_root_clock_control)
+		hws->funcs.dpp_root_clock_control(hws, pipe_ctx->plane_res.dpp->inst, true);
+
 	if (REG(DC_IP_REQUEST_CNTL)) {
 		REG_SET(DC_IP_REQUEST_CNTL, 0,
 				IP_REQUEST_EN, 1);
@ -1158,7 +1162,7 @@ static void dcn20_enable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx,
 	//if (dc->debug.sanity_checks) {
 	//	dcn10_verify_allow_pstate_change_high(dc);
 	//}
-	dcn20_power_on_plane(dc->hwseq, pipe_ctx);
+	dcn20_power_on_plane_resources(dc->hwseq, pipe_ctx);

 	/* enable DCFCLK current DCHUB */
 	pipe_ctx->plane_res.hubp->funcs->hubp_clk_cntl(pipe_ctx->plane_res.hubp, true);
@ -1720,10 +1724,8 @@ static void dcn20_program_pipe(
 				pipe_ctx->pipe_dlg_param.vupdate_offset,
 				pipe_ctx->pipe_dlg_param.vupdate_width);

-		if (pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM) {
-			pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, CRTC_STATE_VBLANK);
+		if (pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM)
 			pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, CRTC_STATE_VACTIVE);
-		}

 		pipe_ctx->stream_res.tg->funcs->set_vtg_params(
 				pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing, true);
@ -1991,6 +1993,16 @@ void dcn20_post_unlock_program_front_end(
 		}
 	}

+	/* P-State support transitions:
+	 * Natural -> FPO: 		P-State disabled in prepare, force disallow anytime is safe
+	 * FPO -> Natural: 		Unforce anytime after FW disable is safe (P-State will assert naturally)
+	 * Unsupported -> FPO:	P-State enabled in optimize, force disallow anytime is safe
+	 * FPO -> Unsupported:	P-State disabled in prepare, unforce disallow anytime is safe
+	 * FPO <-> SubVP:		Force disallow is maintained on the FPO / SubVP pipes
+	 */
+	if (hwseq && hwseq->funcs.update_force_pstate)
+		dc->hwseq->funcs.update_force_pstate(dc, context);
+
 	/* Only program the MALL registers after all the main and phantom pipes
 	 * are done programming.
 	 */
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
@ -712,7 +712,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.timing_trace = false,
 		.clock_trace = true,
 		.disable_pplib_clock_request = true,
-		.pipe_split_policy = MPC_SPLIT_AVOID_MULT_DISP,
+		.pipe_split_policy = MPC_SPLIT_DYNAMIC,
 		.force_single_disp_pipe_split = false,
 		.disable_dcc = DCC_ENABLE,
 		.vsr_support = true,
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@ -621,7 +621,8 @@ void dcn30_init_hw(struct dc *dc)
 	if (dc->clk_mgr->funcs->notify_wm_ranges)
 		dc->clk_mgr->funcs->notify_wm_ranges(dc->clk_mgr);

-	if (dc->clk_mgr->funcs->set_hard_max_memclk)
+	//if softmax is enabled then hardmax will be set by a different call
+	if (dc->clk_mgr->funcs->set_hard_max_memclk && !dc->clk_mgr->dc_mode_softmax_enabled)
 		dc->clk_mgr->funcs->set_hard_max_memclk(dc->clk_mgr);

 	if (dc->res_pool->hubbub->funcs->force_pstate_change_control)
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
@ -291,7 +291,7 @@ static void optc3_set_timing_double_buffer(struct timing_generator *optc, bool e
 		   OTG_DRR_TIMING_DBUF_UPDATE_MODE, mode);
 }

-void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *optc)
+static void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *optc)
 {
 	struct optc *optc1 = DCN10TG_FROM_TG(optc);

--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
@ -2016,6 +2016,8 @@ bool dcn30_can_support_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc,
 	if (context->streams[0]->vrr_active_variable)
 		return false;

+	context->streams[0]->fpo_in_use = true;
+
 	return true;
 }

--- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
@ -687,7 +687,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_clock_gate = true,
 	.disable_pplib_clock_request = true,
 	.disable_pplib_wm_range = true,
-	.pipe_split_policy = MPC_SPLIT_AVOID,
+	.pipe_split_policy = MPC_SPLIT_DYNAMIC,
 	.force_single_disp_pipe_split = false,
 	.disable_dcc = DCC_ENABLE,
 	.vsr_support = true,
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
@ -66,17 +66,8 @@ void dccg31_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk)
 		REG_UPDATE(DPPCLK_DTO_CTRL,
 				DPPCLK_DTO_ENABLE[dpp_inst], 1);
 	} else {
-		//DTO must be enabled to generate a 0Hz clock output
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp) {
-			REG_UPDATE(DPPCLK_DTO_CTRL,
-					DPPCLK_DTO_ENABLE[dpp_inst], 1);
-			REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0,
-					DPPCLK0_DTO_PHASE, 0,
-					DPPCLK0_DTO_MODULO, 1);
-		} else {
-			REG_UPDATE(DPPCLK_DTO_CTRL,
-					DPPCLK_DTO_ENABLE[dpp_inst], 0);
-		}
+		REG_UPDATE(DPPCLK_DTO_CTRL,
+				DPPCLK_DTO_ENABLE[dpp_inst], 0);
 	}
 	dccg->pipe_dppclk_khz[dpp_inst] = req_dppclk;
 }
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
@ -285,7 +285,7 @@ void dcn31_init_hw(struct dc *dc)
 	if (dc->clk_mgr->funcs->notify_wm_ranges)
 		dc->clk_mgr->funcs->notify_wm_ranges(dc->clk_mgr);

-	if (dc->clk_mgr->funcs->set_hard_max_memclk)
+	if (dc->clk_mgr->funcs->set_hard_max_memclk && !dc->clk_mgr->dc_mode_softmax_enabled)
 		dc->clk_mgr->funcs->set_hard_max_memclk(dc->clk_mgr);

 	if (dc->res_pool->hubbub->funcs->force_pstate_change_control)
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
@ -289,8 +289,31 @@ static void dccg314_set_valid_pixel_rate(
 	dccg314_set_dtbclk_dto(dccg, &dto_params);
 }

+static void dccg314_dpp_root_clock_control(
+		struct dccg *dccg,
+		unsigned int dpp_inst,
+		bool clock_on)
+{
+	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+
+	if (clock_on) {
+		/* turn off the DTO and leave phase/modulo at max */
+		REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 0);
+		REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0,
+			  DPPCLK0_DTO_PHASE, 0xFF,
+			  DPPCLK0_DTO_MODULO, 0xFF);
+	} else {
+		/* turn on the DTO to generate a 0hz clock */
+		REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 1);
+		REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0,
+			  DPPCLK0_DTO_PHASE, 0,
+			  DPPCLK0_DTO_MODULO, 1);
+	}
+}
+
 static const struct dccg_funcs dccg314_funcs = {
 	.update_dpp_dto = dccg31_update_dpp_dto,
+	.dpp_root_clock_control = dccg314_dpp_root_clock_control,
 	.get_dccg_ref_freq = dccg31_get_dccg_ref_freq,
 	.dccg_init = dccg31_init,
 	.set_dpstreamclk = dccg314_set_dpstreamclk,
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@ -390,6 +390,16 @@ void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
 				pix_per_cycle);
 }

+void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on)
+{
+	if (!hws->ctx->dc->debug.root_clock_optimization.bits.dpp)
+		return;
+
+	if (hws->ctx->dc->res_pool->dccg->funcs->dpp_root_clock_control)
+		hws->ctx->dc->res_pool->dccg->funcs->dpp_root_clock_control(
+			hws->ctx->dc->res_pool->dccg, dpp_inst, clock_on);
+}
+
 void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on)
 {
 	struct dc_context *ctx = hws->ctx;
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
@ -43,4 +43,6 @@ void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx);

 void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on);

+void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on);
+
 #endif /* __DC_HWSS_DCN314_H__ */
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
@ -137,6 +137,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = {
 	.plane_atomic_disable = dcn20_plane_atomic_disable,
 	.plane_atomic_power_down = dcn10_plane_atomic_power_down,
 	.enable_power_gating_plane = dcn314_enable_power_gating_plane,
+	.dpp_root_clock_control = dcn314_dpp_root_clock_control,
 	.hubp_pg_control = dcn314_hubp_pg_control,
 	.program_all_writeback_pipes_in_tree = dcn30_program_all_writeback_pipes_in_tree,
 	.update_odm = dcn314_update_odm,
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
@ -42,6 +42,20 @@
 #define DC_LOGGER \
 	dccg->ctx->logger

+/* This function is a workaround for writing to OTG_PIXEL_RATE_DIV
+ * without the probability of causing a DIG FIFO error.
+ */
+static void dccg32_wait_for_dentist_change_done(
+	struct dccg *dccg)
+{
+	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+
+	uint32_t dentist_dispclk_value = REG_READ(DENTIST_DISPCLK_CNTL);
+
+	REG_WRITE(DENTIST_DISPCLK_CNTL, dentist_dispclk_value);
+	REG_WAIT(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, 1, 50, 2000);
+}
+
 static void dccg32_get_pixel_rate_div(
 		struct dccg *dccg,
 		uint32_t otg_inst,
@ -110,21 +124,29 @@ static void dccg32_set_pixel_rate_div(
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG0_PIXEL_RATE_DIVK1, k1,
 				OTG0_PIXEL_RATE_DIVK2, k2);
+
+		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	case 1:
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG1_PIXEL_RATE_DIVK1, k1,
 				OTG1_PIXEL_RATE_DIVK2, k2);
+
+		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	case 2:
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG2_PIXEL_RATE_DIVK1, k1,
 				OTG2_PIXEL_RATE_DIVK2, k2);
+
+		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	case 3:
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG3_PIXEL_RATE_DIVK1, k1,
 				OTG3_PIXEL_RATE_DIVK2, k2);
+
+		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	default:
 		BREAK_TO_DEBUGGER();
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h
@ -111,7 +111,8 @@
 	DCCG_SF(DTBCLK_P_CNTL, DTBCLK_P3_SRC_SEL, mask_sh),\
 	DCCG_SF(DTBCLK_P_CNTL, DTBCLK_P3_EN, mask_sh),\
 	DCCG_SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO_SEL, mask_sh),\
-	DCCG_SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO0_SOURCE_SEL, mask_sh)
+	DCCG_SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO0_SOURCE_SEL, mask_sh),\
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh)


 struct dccg *dccg32_create(
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.c
@ -274,10 +274,10 @@ static bool is_dp_dig_pixel_rate_div_policy(struct dc *dc, const struct dc_crtc_
 		dc->debug.enable_dp_dig_pixel_rate_div_policy;
 }

-static void enc32_stream_encoder_dp_unblank(
-        struct dc_link *link,
-		struct stream_encoder *enc,
-		const struct encoder_unblank_param *param)
+void enc32_stream_encoder_dp_unblank(
+	struct dc_link *link,
+	struct stream_encoder *enc,
+	const struct encoder_unblank_param *param)
 {
 	struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc);
 	struct dc *dc = enc->ctx->dc;
@ -286,6 +286,7 @@ static void enc32_stream_encoder_dp_unblank(
 		uint32_t n_vid = 0x8000;
 		uint32_t m_vid;
 		uint32_t n_multiply = 0;
+		uint32_t pix_per_cycle = 0;
 		uint64_t m_vid_l = n_vid;

 		/* YCbCr 4:2:0 : Computed VID_M will be 2X the input rate */
@ -293,6 +294,7 @@ static void enc32_stream_encoder_dp_unblank(
 			|| is_dp_dig_pixel_rate_div_policy(dc, &param->timing)) {
 			/*this logic should be the same in get_pixel_clock_parameters() */
 			n_multiply = 1;
+			pix_per_cycle = 1;
 		}
 		/* M / N = Fstream / Flink
 		 * m_vid / n_vid = pixel rate / link rate
@ -320,6 +322,10 @@ static void enc32_stream_encoder_dp_unblank(
 		REG_UPDATE_2(DP_VID_TIMING,
 				DP_VID_M_N_GEN_EN, 1,
 				DP_VID_N_MUL, n_multiply);
+
+		REG_UPDATE(DP_PIXEL_FORMAT,
+				DP_PIXEL_PER_CYCLE_PROCESSING_MODE,
+				pix_per_cycle);
 	}

 	/* make sure stream is disabled before resetting steer fifo */
@ -434,7 +440,7 @@ static void enc32_reset_fifo(struct stream_encoder *enc, bool reset)
 		udelay(10);
 }

-static void enc32_enable_fifo(struct stream_encoder *enc)
+void enc32_enable_fifo(struct stream_encoder *enc)
 {
 	struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc);

--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.h
@ -194,4 +194,12 @@ void dcn32_dio_stream_encoder_construct(
 	const struct dcn10_stream_encoder_shift *se_shift,
 	const struct dcn10_stream_encoder_mask *se_mask);

+
+void enc32_enable_fifo(struct stream_encoder *enc);
+
+void enc32_stream_encoder_dp_unblank(
+		struct dc_link *link,
+		struct stream_encoder *enc,
+		const struct encoder_unblank_param *param);
+
 #endif /* __DC_DIO_STREAM_ENCODER_DCN32_H__ */
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubp.c
@ -47,6 +47,15 @@ void hubp32_update_force_pstate_disallow(struct hubp *hubp, bool pstate_disallow
 			DATA_UCLK_PSTATE_FORCE_VALUE, 0);
 }

+void hubp32_update_force_cursor_pstate_disallow(struct hubp *hubp, bool pstate_disallow)
+{
+	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
+
+	REG_UPDATE_2(UCLK_PSTATE_FORCE,
+			CURSOR_UCLK_PSTATE_FORCE_EN, pstate_disallow,
+			CURSOR_UCLK_PSTATE_FORCE_VALUE, 0);
+}
+
 void hubp32_update_mall_sel(struct hubp *hubp, uint32_t mall_sel, bool c_cursor)
 {
 	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
@ -188,6 +197,7 @@ static struct hubp_funcs dcn32_hubp_funcs = {
 	.hubp_set_flip_int = hubp1_set_flip_int,
 	.hubp_in_blank = hubp1_in_blank,
 	.hubp_update_force_pstate_disallow = hubp32_update_force_pstate_disallow,
+	.hubp_update_force_cursor_pstate_disallow = hubp32_update_force_cursor_pstate_disallow,
 	.phantom_hubp_post_enable = hubp32_phantom_hubp_post_enable,
 	.hubp_update_mall_sel = hubp32_update_mall_sel,
 	.hubp_prepare_subvp_buffering = hubp32_prepare_subvp_buffering
--- a/Show More
+++ b/Show More