Merge tag 'spi-v5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi

Pull spi updates from Mark Brown:
 "This has been a very active release for the DesignWare driver in
  particular - after a long period of inactivity we have had a lot of
  people actively working on it for unrelated reasons this cycle with
  some of that work still not landed.

  Otherwise it's been fairly quiet for the subsystem.

  Highlights include:

   - Lots of performance improvements and fixes for the DesignWare
     driver from Serge Semin, Andy Shevchenko, Wan Ahmad Zainie, Clement
     Leger, Dinh Nguyen and Jarkko Nikula.

   - Support for octal mode transfers in spidev.

   - Slave mode support for the Rockchip drivers.

   - Support for AMD controllers, Broadcom mspi and Raspberry Pi 4, and
     Intel Elkhart Lake"

* tag 'spi-v5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (125 commits)
  spi: spi-fsl-dspi: fix native data copy
  spi: Convert DW SPI binding to DT schema
  spi: dw: Refactor mid_spi_dma_setup() to separate DMA and IRQ config
  spi: dw: Make DMA request line assignments explicit for Intel Medfield
  spi: bcm2835: Remove shared interrupt support
  dt-bindings: snps,dw-apb-ssi: add optional reset property
  spi: dw: add reset control
  spi: bcm2835: Enable shared interrupt support
  spi: bcm2835: Implement shutdown callback
  spi: dw: Use regset32 DebugFS method to create regdump file
  spi: dw: Add DMA support to the DW SPI MMIO driver
  spi: dw: Cleanup generic DW DMA code namings
  spi: dw: Add DW SPI DMA/PCI/MMIO dependency on the DW SPI core
  spi: dw: Remove DW DMA code dependency from DW_DMAC_PCI
  spi: dw: Move Non-DMA code to the DW PCIe-SPI driver
  spi: dw: Add core suffix to the DW APB SSI core source file
  spi: dw: Fix Rx-only DMA transfers
  spi: dw: Use DMA max burst to set the request thresholds
  spi: dw: Parameterize the DMA Rx/Tx burst length
  spi: dw: Add SPI Rx-done wait method to DMA-based transfer
  ...

55 files changed, 2082 insertions, 941 deletions

diff --git a/Documentation/devicetree/bindings/spi/brcm,spi-bcm-qspi.txt b/Documentation/devicetree/bindings/spi/brcm,spi-bcm-qspi.txt
index ad7ac80a3841..f5e518d099f2 100644
--- a/Documentation/devicetree/bindings/spi/brcm,spi-bcm-qspi.txt
+++ b/Documentation/devicetree/bindings/spi/brcm,spi-bcm-qspi.txt
@@ -26,6 +26,16 @@ Required properties:
     "brcm,spi-bcm-qspi", "brcm,spi-brcmstb-qspi" : MSPI+BSPI on BRCMSTB SoCs
     "brcm,spi-bcm-qspi", "brcm,spi-brcmstb-mspi" : Second Instance of MSPI
 						   BRCMSTB  SoCs
+    "brcm,spi-bcm7425-qspi", "brcm,spi-bcm-qspi", "brcm,spi-brcmstb-mspi" : Second Instance of MSPI
+    			     			  			    BRCMSTB  SoCs
+    "brcm,spi-bcm7429-qspi", "brcm,spi-bcm-qspi", "brcm,spi-brcmstb-mspi" : Second Instance of MSPI
+    			     			  			    BRCMSTB  SoCs
+    "brcm,spi-bcm7435-qspi", "brcm,spi-bcm-qspi", "brcm,spi-brcmstb-mspi" : Second Instance of MSPI
+    			     			  			    BRCMSTB  SoCs
+    "brcm,spi-bcm7216-qspi", "brcm,spi-bcm-qspi", "brcm,spi-brcmstb-mspi" : Second Instance of MSPI
+    			     			  			    BRCMSTB  SoCs
+    "brcm,spi-bcm7278-qspi", "brcm,spi-bcm-qspi", "brcm,spi-brcmstb-mspi" : Second Instance of MSPI
+    			     			  			    BRCMSTB  SoCs
     "brcm,spi-bcm-qspi", "brcm,spi-nsp-qspi"     : MSPI+BSPI on Cygnus, NSP
     "brcm,spi-bcm-qspi", "brcm,spi-ns2-qspi"     : NS2 SoCs
 
diff --git a/Documentation/devicetree/bindings/spi/mikrotik,rb4xx-spi.yaml b/Documentation/devicetree/bindings/spi/mikrotik,rb4xx-spi.yaml
new file mode 100644
index 000000000000..4ddb42a4ae05
--- /dev/null
+++ b/Documentation/devicetree/bindings/spi/mikrotik,rb4xx-spi.yaml
@@ -0,0 +1,36 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/spi/mikrotik,rb4xx-spi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: MikroTik RB4xx series SPI master
+
+maintainers:
+  - Gabor Juhos <juhosg@openwrt.org>
+  - Bert Vermeulen <bert@biot.com>
+
+allOf:
+  - $ref: "spi-controller.yaml#"
+
+properties:
+  compatible:
+    const: mikrotik,rb4xx-spi
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+
+examples:
+  - |
+    spi: spi@1f000000 {
+        #address-cells = <1>;
+        #size-cells = <0>;
+        compatible = "mikrotik,rb4xx-spi";
+        reg = <0x1f000000 0x10>;
+    };
+
+...
\ No newline at end of file
diff --git a/Documentation/devicetree/bindings/spi/renesas,rspi.yaml b/Documentation/devicetree/bindings/spi/renesas,rspi.yaml
new file mode 100644
index 000000000000..c54ac059043f
--- /dev/null
+++ b/Documentation/devicetree/bindings/spi/renesas,rspi.yaml
@@ -0,0 +1,144 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/spi/renesas,rspi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Renesas (Quad) Serial Peripheral Interface (RSPI/QSPI)
+
+maintainers:
+  - Geert Uytterhoeven <geert+renesas@glider.be>
+
+properties:
+  compatible:
+    oneOf:
+      - items:
+          - enum:
+              - renesas,rspi-sh7757    # SH7757
+          - const: renesas,rspi        # Legacy SH
+
+      - items:
+          - enum:
+              - renesas,rspi-r7s72100  # RZ/A1H
+              - renesas,rspi-r7s9210   # RZ/A2
+          - const: renesas,rspi-rz     # RZ/A
+
+      - items:
+          - enum:
+              - renesas,qspi-r8a7743   # RZ/G1M
+              - renesas,qspi-r8a7744   # RZ/G1N
+              - renesas,qspi-r8a7745   # RZ/G1E
+              - renesas,qspi-r8a77470  # RZ/G1C
+              - renesas,qspi-r8a7790   # R-Car H2
+              - renesas,qspi-r8a7791   # R-Car M2-W
+              - renesas,qspi-r8a7792   # R-Car V2H
+              - renesas,qspi-r8a7793   # R-Car M2-N
+              - renesas,qspi-r8a7794   # R-Car E2
+          - const: renesas,qspi        # R-Car Gen2 and RZ/G1
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    oneOf:
+      - items:
+          - description: A combined interrupt
+      - items:
+          - description: Error interrupt (SPEI)
+          - description: Receive Interrupt (SPRI)
+          - description: Transmit Interrupt (SPTI)
+
+  interrupt-names:
+    oneOf:
+      - items:
+          - const: mux
+      - items:
+          - const: error
+          - const: rx
+          - const: tx
+
+  clocks:
+    maxItems: 1
+
+  power-domains:
+    maxItems: 1
+
+  resets:
+    maxItems: 1
+
+  dmas:
+    description:
+      Must contain a list of pairs of references to DMA specifiers, one for
+      transmission, and one for reception.
+
+  dma-names:
+    minItems: 2
+    maxItems: 4
+    items:
+      enum:
+        - tx
+        - rx
+
+  num-cs:
+    description: |
+      Total number of native chip selects.
+      Hardware limitations related to chip selects:
+        - When using GPIO chip selects, at least one native chip select must
+          be left unused, as it will be driven anyway.
+    minimum: 1
+    maximum: 2
+    default: 1
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - clocks
+  - power-domains
+  - '#address-cells'
+  - '#size-cells'
+
+allOf:
+  - $ref: spi-controller.yaml#
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - renesas,rspi-rz
+    then:
+      properties:
+        interrupts:
+          minItems: 3
+      required:
+        - interrupt-names
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - renesas,qspi
+    then:
+      required:
+        - resets
+
+examples:
+  - |
+    #include <dt-bindings/clock/r8a7791-cpg-mssr.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/power/r8a7791-sysc.h>
+
+    qspi: spi@e6b10000 {
+            compatible = "renesas,qspi-r8a7791", "renesas,qspi";
+            reg = <0xe6b10000 0x2c>;
+            interrupts = <GIC_SPI 184 IRQ_TYPE_LEVEL_HIGH>;
+            clocks = <&cpg CPG_MOD 917>;
+            dmas = <&dmac0 0x17>, <&dmac0 0x18>, <&dmac1 0x17>, <&dmac1 0x18>;
+            dma-names = "tx", "rx", "tx", "rx";
+            power-domains = <&sysc R8A7791_PD_ALWAYS_ON>;
+            resets = <&cpg 917>;
+            num-cs = <1>;
+            #address-cells = <1>;
+            #size-cells = <0>;
+    };
diff --git a/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.txt b/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.txt
deleted file mode 100644
index 3ed08ee9feba..000000000000
--- a/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.txt
+++ /dev/null
@@ -1,41 +0,0 @@
-Synopsys DesignWare AMBA 2.0 Synchronous Serial Interface.
-
-Required properties:
-- compatible : "snps,dw-apb-ssi" or "mscc,<soc>-spi", where soc is "ocelot" or
-  "jaguar2", or "amazon,alpine-dw-apb-ssi"
-- reg : The register base for the controller. For "mscc,<soc>-spi", a second
-  register set is required (named ICPU_CFG:SPI_MST)
-- interrupts : One interrupt, used by the controller.
-- #address-cells : <1>, as required by generic SPI binding.
-- #size-cells : <0>, also as required by generic SPI binding.
-- clocks : phandles for the clocks, see the description of clock-names below.
-   The phandle for the "ssi_clk" is required. The phandle for the "pclk" clock
-   is optional. If a single clock is specified but no clock-name, it is the
-   "ssi_clk" clock. If both clocks are listed, the "ssi_clk" must be first.
-
-Optional properties:
-- clock-names : Contains the names of the clocks:
-    "ssi_clk", for the core clock used to generate the external SPI clock.
-    "pclk", the interface clock, required for register access. If a clock domain
-     used to enable this clock then it should be named "pclk_clkdomain".
-- cs-gpios : Specifies the gpio pins to be used for chipselects.
-- num-cs : The number of chipselects. If omitted, this will default to 4.
-- reg-io-width : The I/O register width (in bytes) implemented by this
-  device.  Supported values are 2 or 4 (the default).
-
-Child nodes as per the generic SPI binding.
-
-Example:
-
-	spi@fff00000 {
-		compatible = "snps,dw-apb-ssi";
-		reg = <0xfff00000 0x1000>;
-		interrupts = <0 154 4>;
-		#address-cells = <1>;
-		#size-cells = <0>;
-		clocks = <&spi_m_clk>;
-		num-cs = <2>;
-		cs-gpios = <&gpio0 13 0>,
-			   <&gpio0 14 0>;
-	};
-
diff --git a/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml b/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml
new file mode 100644
index 000000000000..c62cbe79f00d
--- /dev/null
+++ b/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml
@@ -0,0 +1,133 @@
+# SPDX-License-Identifier: GPL-2.0-only
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/spi/snps,dw-apb-ssi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Synopsys DesignWare AMBA 2.0 Synchronous Serial Interface
+
+maintainers:
+  - Mark Brown <broonie@kernel.org>
+
+allOf:
+  - $ref: "spi-controller.yaml#"
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - mscc,ocelot-spi
+              - mscc,jaguar2-spi
+    then:
+      properties:
+        reg:
+          minItems: 2
+
+properties:
+  compatible:
+    oneOf:
+      - description: Generic DW SPI Controller
+        enum:
+          - snps,dw-apb-ssi
+          - snps,dwc-ssi-1.01a
+      - description: Microsemi Ocelot/Jaguar2 SoC SPI Controller
+        items:
+          - enum:
+              - mscc,ocelot-spi
+              - mscc,jaguar2-spi
+          - const: snps,dw-apb-ssi
+      - description: Amazon Alpine SPI Controller
+        const: amazon,alpine-dw-apb-ssi
+      - description: Renesas RZ/N1 SPI Controller
+        items:
+          - const: renesas,rzn1-spi
+          - const: snps,dw-apb-ssi
+      - description: Intel Keem Bay SPI Controller
+        const: intel,keembay-ssi
+
+  reg:
+    minItems: 1
+    items:
+      - description: DW APB SSI controller memory mapped registers
+      - description: SPI MST region map
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    minItems: 1
+    items:
+      - description: SPI Controller reference clock source
+      - description: APB interface clock source
+
+  clock-names:
+    minItems: 1
+    items:
+      - const: ssi_clk
+      - const: pclk
+
+  resets:
+    maxItems: 1
+
+  reset-names:
+    const: spi
+
+  reg-io-width:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: I/O register width (in bytes) implemented by this device
+    default: 4
+    enum: [ 2, 4 ]
+
+  num-cs:
+    default: 4
+    minimum: 1
+    maximum: 4
+
+  dmas:
+    items:
+      - description: TX DMA Channel
+      - description: RX DMA Channel
+
+  dma-names:
+    items:
+      - const: tx
+      - const: rx
+
+patternProperties:
+  "^.*@[0-9a-f]+$":
+    type: object
+    properties:
+      reg:
+        minimum: 0
+        maximum: 3
+
+      spi-rx-bus-width:
+        const: 1
+
+      spi-tx-bus-width:
+        const: 1
+
+unevaluatedProperties: false
+
+required:
+  - compatible
+  - reg
+  - "#address-cells"
+  - "#size-cells"
+  - interrupts
+  - clocks
+
+examples:
+  - |
+    spi@fff00000 {
+      compatible = "snps,dw-apb-ssi";
+      reg = <0xfff00000 0x1000>;
+      #address-cells = <1>;
+      #size-cells = <0>;
+      interrupts = <0 154 4>;
+      clocks = <&spi_m_clk>;
+      num-cs = <2>;
+      cs-gpios = <&gpio0 13 0>,
+                 <&gpio0 14 0>;
+    };
+...
diff --git a/Documentation/devicetree/bindings/spi/socionext,uniphier-spi.yaml b/Documentation/devicetree/bindings/spi/socionext,uniphier-spi.yaml
new file mode 100644
index 000000000000..c25409298bdf
--- /dev/null
+++ b/Documentation/devicetree/bindings/spi/socionext,uniphier-spi.yaml
@@ -0,0 +1,57 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/spi/socionext,uniphier-spi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Socionext UniPhier SPI controller
+
+description: |
+  UniPhier SoCs have SCSSI which supports SPI single channel.
+
+maintainers:
+  - Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
+  - Keiji Hayashibara <hayashibara.keiji@socionext.com>
+
+allOf:
+  - $ref: spi-controller.yaml#
+
+properties:
+  "#address-cells": true
+  "#size-cells": true
+
+  compatible:
+    const: socionext,uniphier-scssi
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    maxItems: 1
+
+  resets:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - clocks
+  - resets
+  - "#address-cells"
+  - "#size-cells"
+
+examples:
+  - |
+    spi0: spi@54006000 {
+        compatible = "socionext,uniphier-scssi";
+        reg = <0x54006000 0x100>;
+        #address-cells = <1>;
+        #size-cells = <0>;
+        interrupts = <0 39 4>;
+        clocks = <&peri_clk 11>;
+        resets = <&peri_rst 11>;
+    };
diff --git a/Documentation/devicetree/bindings/spi/spi-dw.txt b/Documentation/devicetree/bindings/spi/spi-dw.txt
deleted file mode 100644
index 7b63ed601990..000000000000
--- a/Documentation/devicetree/bindings/spi/spi-dw.txt
+++ /dev/null
@@ -1,24 +0,0 @@
-Synopsys DesignWare SPI master
-
-Required properties:
-- compatible: should be "snps,designware-spi"
-- #address-cells: see spi-bus.txt
-- #size-cells: see spi-bus.txt
-- reg: address and length of the spi master registers
-- interrupts: should contain one interrupt
-- clocks: spi clock phandle
-- num-cs: see spi-bus.txt
-
-Optional properties:
-- cs-gpios: see spi-bus.txt
-
-Example:
-
-spi: spi@4020a000 {
-	compatible = "snps,designware-spi";
-	interrupts = <11 1>;
-	reg = <0x4020a000 0x1000>;
-	clocks = <&pclk>;
-	num-cs = <2>;
-	cs-gpios = <&banka 0 0>;
-};
diff --git a/Documentation/devicetree/bindings/spi/spi-rspi.txt b/Documentation/devicetree/bindings/spi/spi-rspi.txt
deleted file mode 100644
index 421722b93992..000000000000
--- a/Documentation/devicetree/bindings/spi/spi-rspi.txt
+++ /dev/null
@@ -1,73 +0,0 @@
-Device tree configuration for Renesas RSPI/QSPI driver
-
-Required properties:
-- compatible       : For Renesas Serial Peripheral Interface on legacy SH:
-		     "renesas,rspi-<soctype>", "renesas,rspi" as fallback.
-		     For Renesas Serial Peripheral Interface on RZ/A:
-		     "renesas,rspi-<soctype>", "renesas,rspi-rz" as fallback.
-		     For Quad Serial Peripheral Interface on R-Car Gen2 and
-		     RZ/G1 devices:
-		     "renesas,qspi-<soctype>", "renesas,qspi" as fallback.
-		     Examples with soctypes are:
-		        - "renesas,rspi-sh7757" (SH)
-			- "renesas,rspi-r7s72100" (RZ/A1H)
-			- "renesas,rspi-r7s9210" (RZ/A2)
-			- "renesas,qspi-r8a7743" (RZ/G1M)
-			- "renesas,qspi-r8a7744" (RZ/G1N)
-			- "renesas,qspi-r8a7745" (RZ/G1E)
-			- "renesas,qspi-r8a77470" (RZ/G1C)
-			- "renesas,qspi-r8a7790" (R-Car H2)
-			- "renesas,qspi-r8a7791" (R-Car M2-W)
-			- "renesas,qspi-r8a7792" (R-Car V2H)
-			- "renesas,qspi-r8a7793" (R-Car M2-N)
-			- "renesas,qspi-r8a7794" (R-Car E2)
-- reg              : Address start and address range size of the device
-- interrupts       : A list of interrupt-specifiers, one for each entry in
-		     interrupt-names.
-		     If interrupt-names is not present, an interrupt specifier
-		     for a single muxed interrupt.
-- interrupt-names  : A list of interrupt names. Should contain (if present):
-		       - "error" for SPEI,
-		       - "rx" for SPRI,
-		       - "tx" to SPTI,
-		       - "mux" for a single muxed interrupt.
-- num-cs	   : Number of chip selects. Some RSPI cores have more than 1.
-- #address-cells   : Must be <1>
-- #size-cells      : Must be <0>
-
-Optional properties:
-- clocks           : Must contain a reference to the functional clock.
-- dmas             : Must contain a list of two references to DMA specifiers,
-		     one for transmission, and one for reception.
-- dma-names        : Must contain a list of two DMA names, "tx" and "rx".
-
-Pinctrl properties might be needed, too.  See
-Documentation/devicetree/bindings/pinctrl/renesas,*.
-
-Examples:
-
-	spi0: spi@e800c800 {
-		compatible = "renesas,rspi-r7s72100", "renesas,rspi-rz";
-		reg = <0xe800c800 0x24>;
-		interrupts = <0 238 IRQ_TYPE_LEVEL_HIGH>,
-			     <0 239 IRQ_TYPE_LEVEL_HIGH>,
-			     <0 240 IRQ_TYPE_LEVEL_HIGH>;
-		interrupt-names = "error", "rx", "tx";
-		interrupt-parent = <&gic>;
-		num-cs = <1>;
-		#address-cells = <1>;
-		#size-cells = <0>;
-	};
-
-	spi: spi@e6b10000 {
-		compatible = "renesas,qspi-r8a7791", "renesas,qspi";
-		reg = <0 0xe6b10000 0 0x2c>;
-		interrupt-parent = <&gic>;
-		interrupts = <0 184 IRQ_TYPE_LEVEL_HIGH>;
-		clocks = <&mstp9_clks R8A7791_CLK_QSPI_MOD>;
-		num-cs = <1>;
-		#address-cells = <1>;
-		#size-cells = <0>;
-		dmas = <&dmac0 0x17>, <&dmac0 0x18>;
-		dma-names = "tx", "rx";
-	};
diff --git a/Documentation/devicetree/bindings/spi/spi-uniphier.txt b/Documentation/devicetree/bindings/spi/spi-uniphier.txt
deleted file mode 100644
index e1201573a29a..000000000000
--- a/Documentation/devicetree/bindings/spi/spi-uniphier.txt
+++ /dev/null
@@ -1,28 +0,0 @@
-Socionext UniPhier SPI controller driver
-
-UniPhier SoCs have SCSSI which supports SPI single channel.
-
-Required properties:
- - compatible: should be "socionext,uniphier-scssi"
- - reg: address and length of the spi master registers
- - #address-cells: must be <1>, see spi-bus.txt
- - #size-cells: must be <0>, see spi-bus.txt
- - interrupts: a single interrupt specifier
- - pinctrl-names: should be "default"
- - pinctrl-0: pin control state for the default mode
- - clocks: a phandle to the clock for the device
- - resets: a phandle to the reset control for the device
-
-Example:
-
-spi0: spi@54006000 {
-	compatible = "socionext,uniphier-scssi";
-	reg = <0x54006000 0x100>;
-	#address-cells = <1>;
-	#size-cells = <0>;
-	interrupts = <0 39 4>;
-	pinctrl-names = "default";
-	pinctrl-0 = <&pinctrl_spi0>;
-	clocks = <&peri_clk 11>;
-	resets = <&peri_rst 11>;
-};
diff --git a/Documentation/devicetree/bindings/spi/ti_qspi.txt b/Documentation/devicetree/bindings/spi/ti_qspi.txt
index e65fde4a7388..47b184bce414 100644
--- a/Documentation/devicetree/bindings/spi/ti_qspi.txt
+++ b/Documentation/devicetree/bindings/spi/ti_qspi.txt
@@ -29,7 +29,7 @@ modification to bootloader.
 Example:
 
 For am4372:
-qspi: qspi@4b300000 {
+qspi: qspi@47900000 {
 	compatible = "ti,am4372-qspi";
 	reg = <0x47900000 0x100>, <0x30000000 0x4000000>;
 	reg-names = "qspi_base", "qspi_mmap";
diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml
index d3891386d671..d3277fe6640b 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.yaml
+++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml
@@ -633,6 +633,8 @@ patternProperties:
     description: Microsoft Corporation
   "^mikroe,.*":
     description: MikroElektronika d.o.o.
+  "^mikrotik,.*":
+    description: MikroTik
   "^miniand,.*":
     description: Miniand Tech
   "^minix,.*":
diff --git a/MAINTAINERS b/MAINTAINERS
index 8ff132775276..6bc4cdb20038 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -899,6 +899,11 @@ F:	drivers/gpu/drm/amd/include/v9_structs.h
 F:	drivers/gpu/drm/amd/include/vi_structs.h
 F:	include/uapi/linux/kfd_ioctl.h
 
+AMD SPI DRIVER
+M:	Sanjay R Mehta <sanju.mehta@amd.com>
+S:	Maintained
+F:	drivers/spi/spi-amd.c
+
 AMD MP2 I2C DRIVER
 M:	Elie Morisse <syniurge@gmail.com>
 M:	Nehal Shah <nehal-bakulchandra.shah@amd.com>
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 741b9140992a..8f1f8fca79e3 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -226,17 +226,20 @@ config SPI_DESIGNWARE
 	help
 	  general driver for SPI controller core from DesignWare
 
+if SPI_DESIGNWARE
+
+config SPI_DW_DMA
+	bool "DMA support for DW SPI controller"
+
 config SPI_DW_PCI
 	tristate "PCI interface driver for DW SPI core"
-	depends on SPI_DESIGNWARE && PCI
-
-config SPI_DW_MID_DMA
-	bool "DMA support for DW SPI controller on Intel MID platform"
-	depends on SPI_DW_PCI && DW_DMAC_PCI
+	depends on PCI
 
 config SPI_DW_MMIO
 	tristate "Memory-mapped io interface driver for DW SPI core"
-	depends on SPI_DESIGNWARE
+	depends on HAS_IOMEM
+
+endif
 
 config SPI_DLN2
        tristate "Diolan DLN-2 USB SPI adapter"
@@ -844,6 +847,7 @@ config SPI_TXX9
 config SPI_UNIPHIER
 	tristate "Socionext UniPhier SPI Controller"
 	depends on (ARCH_UNIPHIER || COMPILE_TEST) && OF
+	depends on HAS_IOMEM
 	help
 	  This enables a driver for the Socionext UniPhier SoC SCSSI SPI controller.
 
@@ -910,6 +914,12 @@ config SPI_ZYNQMP_GQSPI
 	help
 	  Enables Xilinx GQSPI controller driver for Zynq UltraScale+ MPSoC.
 
+config SPI_AMD
+	tristate "AMD SPI controller"
+	depends on SPI_MASTER || COMPILE_TEST
+	help
+	  Enables SPI controller driver for AMD SoC.
+
 #
 # Add new SPI master controllers in alphabetical order above this line
 #
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index 28f601327f8c..d2e41d3d464a 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -36,9 +36,10 @@ obj-$(CONFIG_SPI_COLDFIRE_QSPI)		+= spi-coldfire-qspi.o
 obj-$(CONFIG_SPI_DAVINCI)		+= spi-davinci.o
 obj-$(CONFIG_SPI_DLN2)			+= spi-dln2.o
 obj-$(CONFIG_SPI_DESIGNWARE)		+= spi-dw.o
+spi-dw-y				:= spi-dw-core.o
+spi-dw-$(CONFIG_SPI_DW_DMA)		+= spi-dw-dma.o
 obj-$(CONFIG_SPI_DW_MMIO)		+= spi-dw-mmio.o
-obj-$(CONFIG_SPI_DW_PCI)		+= spi-dw-midpci.o
-spi-dw-midpci-objs			:= spi-dw-pci.o spi-dw-mid.o
+obj-$(CONFIG_SPI_DW_PCI)		+= spi-dw-pci.o
 obj-$(CONFIG_SPI_EFM32)			+= spi-efm32.o
 obj-$(CONFIG_SPI_EP93XX)		+= spi-ep93xx.o
 obj-$(CONFIG_SPI_FALCON)		+= spi-falcon.o
@@ -127,6 +128,7 @@ obj-$(CONFIG_SPI_XLP)			+= spi-xlp.o
 obj-$(CONFIG_SPI_XTENSA_XTFPGA)		+= spi-xtensa-xtfpga.o
 obj-$(CONFIG_SPI_ZYNQ_QSPI)		+= spi-zynq-qspi.o
 obj-$(CONFIG_SPI_ZYNQMP_GQSPI)		+= spi-zynqmp-gqspi.o
+obj-$(CONFIG_SPI_AMD)			+= spi-amd.o
 
 # SPI slave protocol handlers
 obj-$(CONFIG_SPI_SLAVE_TIME)		+= spi-slave-time.o
diff --git a/drivers/spi/spi-amd.c b/drivers/spi/spi-amd.c
new file mode 100644
index 000000000000..d0aacd4de1b9
--- /dev/null
+++ b/drivers/spi/spi-amd.c
@@ -0,0 +1,315 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+//
+// AMD SPI controller driver
+//
+// Copyright (c) 2020, Advanced Micro Devices, Inc.
+//
+// Author: Sanjay R Mehta <sanju.mehta@amd.com>
+
+#include <linux/acpi.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/delay.h>
+#include <linux/spi/spi.h>
+
+#define AMD_SPI_CTRL0_REG	0x00
+#define AMD_SPI_EXEC_CMD	BIT(16)
+#define AMD_SPI_FIFO_CLEAR	BIT(20)
+#define AMD_SPI_BUSY		BIT(31)
+
+#define AMD_SPI_OPCODE_MASK	0xFF
+
+#define AMD_SPI_ALT_CS_REG	0x1D
+#define AMD_SPI_ALT_CS_MASK	0x3
+
+#define AMD_SPI_FIFO_BASE	0x80
+#define AMD_SPI_TX_COUNT_REG	0x48
+#define AMD_SPI_RX_COUNT_REG	0x4B
+#define AMD_SPI_STATUS_REG	0x4C
+
+#define AMD_SPI_MEM_SIZE	200
+
+/* M_CMD OP codes for SPI */
+#define AMD_SPI_XFER_TX		1
+#define AMD_SPI_XFER_RX		2
+
+struct amd_spi {
+	void __iomem *io_remap_addr;
+	unsigned long io_base_addr;
+	u32 rom_addr;
+	u8 chip_select;
+};
+
+static inline u8 amd_spi_readreg8(struct spi_master *master, int idx)
+{
+	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+
+	return ioread8((u8 __iomem *)amd_spi->io_remap_addr + idx);
+}
+
+static inline void amd_spi_writereg8(struct spi_master *master, int idx,
+				     u8 val)
+{
+	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+
+	iowrite8(val, ((u8 __iomem *)amd_spi->io_remap_addr + idx));
+}
+
+static inline void amd_spi_setclear_reg8(struct spi_master *master, int idx,
+					 u8 set, u8 clear)
+{
+	u8 tmp = amd_spi_readreg8(master, idx);
+
+	tmp = (tmp & ~clear) | set;
+	amd_spi_writereg8(master, idx, tmp);
+}
+
+static inline u32 amd_spi_readreg32(struct spi_master *master, int idx)
+{
+	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+
+	return ioread32((u8 __iomem *)amd_spi->io_remap_addr + idx);
+}
+
+static inline void amd_spi_writereg32(struct spi_master *master, int idx,
+				      u32 val)
+{
+	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+
+	iowrite32(val, ((u8 __iomem *)amd_spi->io_remap_addr + idx));
+}
+
+static inline void amd_spi_setclear_reg32(struct spi_master *master, int idx,
+					  u32 set, u32 clear)
+{
+	u32 tmp = amd_spi_readreg32(master, idx);
+
+	tmp = (tmp & ~clear) | set;
+	amd_spi_writereg32(master, idx, tmp);
+}
+
+static void amd_spi_select_chip(struct spi_master *master)
+{
+	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+	u8 chip_select = amd_spi->chip_select;
+
+	amd_spi_setclear_reg8(master, AMD_SPI_ALT_CS_REG, chip_select,
+			      AMD_SPI_ALT_CS_MASK);
+}
+
+static void amd_spi_clear_fifo_ptr(struct spi_master *master)
+{
+	amd_spi_setclear_reg32(master, AMD_SPI_CTRL0_REG, AMD_SPI_FIFO_CLEAR,
+			       AMD_SPI_FIFO_CLEAR);
+}
+
+static void amd_spi_set_opcode(struct spi_master *master, u8 cmd_opcode)
+{
+	amd_spi_setclear_reg32(master, AMD_SPI_CTRL0_REG, cmd_opcode,
+			       AMD_SPI_OPCODE_MASK);
+}
+
+static inline void amd_spi_set_rx_count(struct spi_master *master,
+					u8 rx_count)
+{
+	amd_spi_setclear_reg8(master, AMD_SPI_RX_COUNT_REG, rx_count, 0xff);
+}
+
+static inline void amd_spi_set_tx_count(struct spi_master *master,
+					u8 tx_count)
+{
+	amd_spi_setclear_reg8(master, AMD_SPI_TX_COUNT_REG, tx_count, 0xff);
+}
+
+static inline int amd_spi_busy_wait(struct amd_spi *amd_spi)
+{
+	bool spi_busy;
+	int timeout = 100000;
+
+	/* poll for SPI bus to become idle */
+	spi_busy = (ioread32((u8 __iomem *)amd_spi->io_remap_addr +
+		    AMD_SPI_CTRL0_REG) & AMD_SPI_BUSY) == AMD_SPI_BUSY;
+	while (spi_busy) {
+		usleep_range(10, 20);
+		if (timeout-- < 0)
+			return -ETIMEDOUT;
+
+		spi_busy = (ioread32((u8 __iomem *)amd_spi->io_remap_addr +
+			    AMD_SPI_CTRL0_REG) & AMD_SPI_BUSY) == AMD_SPI_BUSY;
+	}
+
+	return 0;
+}
+
+static void amd_spi_execute_opcode(struct spi_master *master)
+{
+	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+
+	/* Set ExecuteOpCode bit in the CTRL0 register */
+	amd_spi_setclear_reg32(master, AMD_SPI_CTRL0_REG, AMD_SPI_EXEC_CMD,
+			       AMD_SPI_EXEC_CMD);
+
+	amd_spi_busy_wait(amd_spi);
+}
+
+static int amd_spi_master_setup(struct spi_device *spi)
+{
+	struct spi_master *master = spi->master;
+
+	amd_spi_clear_fifo_ptr(master);
+
+	return 0;
+}
+
+static inline int amd_spi_fifo_xfer(struct amd_spi *amd_spi,
+				    struct spi_master *master,
+				    struct spi_message *message)
+{
+	struct spi_transfer *xfer = NULL;
+	u8 cmd_opcode;
+	u8 *buf = NULL;
+	u32 m_cmd = 0;
+	u32 i = 0;
+	u32 tx_len = 0, rx_len = 0;
+
+	list_for_each_entry(xfer, &message->transfers,
+			    transfer_list) {
+		if (xfer->rx_buf)
+			m_cmd = AMD_SPI_XFER_RX;
+		if (xfer->tx_buf)
+			m_cmd = AMD_SPI_XFER_TX;
+
+		if (m_cmd & AMD_SPI_XFER_TX) {
+			buf = (u8 *)xfer->tx_buf;
+			tx_len = xfer->len - 1;
+			cmd_opcode = *(u8 *)xfer->tx_buf;
+			buf++;
+			amd_spi_set_opcode(master, cmd_opcode);
+
+			/* Write data into the FIFO. */
+			for (i = 0; i < tx_len; i++) {
+				iowrite8(buf[i],
+					 ((u8 __iomem *)amd_spi->io_remap_addr +
+					 AMD_SPI_FIFO_BASE + i));
+			}
+
+			amd_spi_set_tx_count(master, tx_len);
+			amd_spi_clear_fifo_ptr(master);
+			/* Execute command */
+			amd_spi_execute_opcode(master);
+		}
+		if (m_cmd & AMD_SPI_XFER_RX) {
+			/*
+			 * Store no. of bytes to be received from
+			 * FIFO
+			 */
+			rx_len = xfer->len;
+			buf = (u8 *)xfer->rx_buf;
+			amd_spi_set_rx_count(master, rx_len);
+			amd_spi_clear_fifo_ptr(master);
+			/* Execute command */
+			amd_spi_execute_opcode(master);
+			/* Read data from FIFO to receive buffer  */
+			for (i = 0; i < rx_len; i++)
+				buf[i] = amd_spi_readreg8(master,
+							  AMD_SPI_FIFO_BASE +
+							  tx_len + i);
+		}
+	}
+
+	/* Update statistics */
+	message->actual_length = tx_len + rx_len + 1;
+	/* complete the transaction */
+	message->status = 0;
+	spi_finalize_current_message(master);
+
+	return 0;
+}
+
+static int amd_spi_master_transfer(struct spi_master *master,
+				   struct spi_message *msg)
+{
+	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+	struct spi_device *spi = msg->spi;
+
+	amd_spi->chip_select = spi->chip_select;
+	amd_spi_select_chip(master);
+
+	/*
+	 * Extract spi_transfers from the spi message and
+	 * program the controller.
+	 */
+	amd_spi_fifo_xfer(amd_spi, master, msg);
+
+	return 0;
+}
+
+static int amd_spi_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct spi_master *master;
+	struct amd_spi *amd_spi;
+	struct resource *res;
+	int err = 0;
+
+	/* Allocate storage for spi_master and driver private data */
+	master = spi_alloc_master(dev, sizeof(struct amd_spi));
+	if (!master) {
+		dev_err(dev, "Error allocating SPI master\n");
+		return -ENOMEM;
+	}
+
+	amd_spi = spi_master_get_devdata(master);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	amd_spi->io_remap_addr = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(amd_spi->io_remap_addr)) {
+		err = PTR_ERR(amd_spi->io_remap_addr);
+		dev_err(dev, "error %d ioremap of SPI registers failed\n", err);
+		goto err_free_master;
+	}
+	dev_dbg(dev, "io_remap_address: %p\n", amd_spi->io_remap_addr);
+
+	/* Initialize the spi_master fields */
+	master->bus_num = 0;
+	master->num_chipselect = 4;
+	master->mode_bits = 0;
+	master->flags = SPI_MASTER_HALF_DUPLEX;
+	master->setup = amd_spi_master_setup;
+	master->transfer_one_message = amd_spi_master_transfer;
+
+	/* Register the controller with SPI framework */
+	err = devm_spi_register_master(dev, master);
+	if (err) {
+		dev_err(dev, "error %d registering SPI controller\n", err);
+		goto err_free_master;
+	}
+
+	return 0;
+
+err_free_master:
+	spi_master_put(master);
+
+	return err;
+}
+
+static const struct acpi_device_id spi_acpi_match[] = {
+	{ "AMDI0061", 0 },
+	{},
+};
+MODULE_DEVICE_TABLE(acpi, spi_acpi_match);
+
+static struct platform_driver amd_spi_driver = {
+	.driver = {
+		.name = "amd_spi",
+		.acpi_match_table = ACPI_PTR(spi_acpi_match),
+	},
+	.probe = amd_spi_probe,
+};
+
+module_platform_driver(amd_spi_driver);
+
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_AUTHOR("Sanjay Mehta <sanju.mehta@amd.com>");
+MODULE_DESCRIPTION("AMD SPI Master Controller Driver");
diff --git a/drivers/spi/spi-armada-3700.c b/drivers/spi/spi-armada-3700.c
index e450ee17787f..fcde419e480c 100644
--- a/drivers/spi/spi-armada-3700.c
+++ b/drivers/spi/spi-armada-3700.c
@@ -276,11 +276,11 @@ static int a3700_spi_fifo_flush(struct a3700_spi *a3700_spi)
 	return -ETIMEDOUT;
 }
 
-static int a3700_spi_init(struct a3700_spi *a3700_spi)
+static void a3700_spi_init(struct a3700_spi *a3700_spi)
 {
 	struct spi_master *master = a3700_spi->master;
 	u32 val;
-	int i, ret = 0;
+	int i;
 
 	/* Reset SPI unit */
 	val = spireg_read(a3700_spi, A3700_SPI_IF_CFG_REG);
@@ -311,8 +311,6 @@ static int a3700_spi_init(struct a3700_spi *a3700_spi)
 	/* Mask the interrupts and clear cause bits */
 	spireg_write(a3700_spi, A3700_SPI_INT_MASK_REG, 0);
 	spireg_write(a3700_spi, A3700_SPI_INT_STAT_REG, ~0U);
-
-	return ret;
 }
 
 static irqreturn_t a3700_spi_interrupt(int irq, void *dev_id)
@@ -886,9 +884,7 @@ static int a3700_spi_probe(struct platform_device *pdev)
 	master->min_speed_hz = DIV_ROUND_UP(clk_get_rate(spi->clk),
 						A3700_SPI_MAX_PRESCALE);
 
-	ret = a3700_spi_init(spi);
-	if (ret)
-		goto error_clk;
+	a3700_spi_init(spi);
 
 	ret = devm_request_irq(dev, spi->irq, a3700_spi_interrupt, 0,
 			       dev_name(dev), master);
diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c
index 013458cabe3c..57ee8c3b7972 100644
--- a/drivers/spi/spi-atmel.c
+++ b/drivers/spi/spi-atmel.c
@@ -706,6 +706,7 @@ static void atmel_spi_next_xfer_pio(struct spi_master *master,
 static int atmel_spi_next_xfer_dma_submit(struct spi_master *master,
 				struct spi_transfer *xfer,
 				u32 *plen)
+	__must_hold(&as->lock)
 {
 	struct atmel_spi	*as = spi_master_get_devdata(master);
 	struct dma_chan		*rxchan = master->dma_rx;
diff --git a/drivers/spi/spi-axi-spi-engine.c b/drivers/spi/spi-axi-spi-engine.c
index eb9b78a90dcf..af86e6d6e16b 100644
--- a/drivers/spi/spi-axi-spi-engine.c
+++ b/drivers/spi/spi-axi-spi-engine.c
@@ -489,22 +489,6 @@ static int spi_engine_probe(struct platform_device *pdev)
 
 	spin_lock_init(&spi_engine->lock);
 
-	spi_engine->base = devm_platform_ioremap_resource(pdev, 0);
-	if (IS_ERR(spi_engine->base)) {
-		ret = PTR_ERR(spi_engine->base);
-		goto err_put_master;
-	}
-
-	version = readl(spi_engine->base + SPI_ENGINE_REG_VERSION);
-	if (SPI_ENGINE_VERSION_MAJOR(version) != 1) {
-		dev_err(&pdev->dev, "Unsupported peripheral version %u.%u.%c\n",
-			SPI_ENGINE_VERSION_MAJOR(version),
-			SPI_ENGINE_VERSION_MINOR(version),
-			SPI_ENGINE_VERSION_PATCH(version));
-		ret = -ENODEV;
-		goto err_put_master;
-	}
-
 	spi_engine->clk = devm_clk_get(&pdev->dev, "s_axi_aclk");
 	if (IS_ERR(spi_engine->clk)) {
 		ret = PTR_ERR(spi_engine->clk);
@@ -525,6 +509,22 @@ static int spi_engine_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_clk_disable;
 
+	spi_engine->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(spi_engine->base)) {
+		ret = PTR_ERR(spi_engine->base);
+		goto err_ref_clk_disable;
+	}
+
+	version = readl(spi_engine->base + SPI_ENGINE_REG_VERSION);
+	if (SPI_ENGINE_VERSION_MAJOR(version) != 1) {
+		dev_err(&pdev->dev, "Unsupported peripheral version %u.%u.%c\n",
+			SPI_ENGINE_VERSION_MAJOR(version),
+			SPI_ENGINE_VERSION_MINOR(version),
+			SPI_ENGINE_VERSION_PATCH(version));
+		ret = -ENODEV;
+		goto err_ref_clk_disable;
+	}
+
 	writel_relaxed(0x00, spi_engine->base + SPI_ENGINE_REG_RESET);
 	writel_relaxed(0xff, spi_engine->base + SPI_ENGINE_REG_INT_PENDING);
 	writel_relaxed(0x00, spi_engine->base + SPI_ENGINE_REG_INT_ENABLE);
diff --git a/drivers/spi/spi-bcm-qspi.c b/drivers/spi/spi-bcm-qspi.c
index 23d295f36c80..681d09085175 100644
--- a/drivers/spi/spi-bcm-qspi.c
+++ b/drivers/spi/spi-bcm-qspi.c
@@ -91,6 +91,7 @@
 #define MSPI_MSPI_STATUS			0x020
 #define MSPI_CPTQP				0x024
 #define MSPI_SPCR3				0x028
+#define MSPI_REV				0x02c
 #define MSPI_TXRAM				0x040
 #define MSPI_RXRAM				0x0c0
 #define MSPI_CDRAM				0x140
@@ -106,14 +107,22 @@
 #define MSPI_SPCR2_SPE				BIT(6)
 #define MSPI_SPCR2_CONT_AFTER_CMD		BIT(7)
 
+#define MSPI_SPCR3_FASTBR			BIT(0)
+#define MSPI_SPCR3_FASTDT			BIT(1)
+#define MSPI_SPCR3_SYSCLKSEL_MASK		GENMASK(11, 10)
+#define MSPI_SPCR3_SYSCLKSEL_27			(MSPI_SPCR3_SYSCLKSEL_MASK & \
+						 ~(BIT(10) | BIT(11)))
+#define MSPI_SPCR3_SYSCLKSEL_108		(MSPI_SPCR3_SYSCLKSEL_MASK & \
+						 BIT(11))
+
 #define MSPI_MSPI_STATUS_SPIF			BIT(0)
 
 #define INTR_BASE_BIT_SHIFT			0x02
 #define INTR_COUNT				0x07
 
 #define NUM_CHIPSELECT				4
-#define QSPI_SPBR_MIN				8U
 #define QSPI_SPBR_MAX				255U
+#define MSPI_BASE_FREQ				27000000UL
 
 #define OPCODE_DIOR				0xBB
 #define OPCODE_QIOR				0xEB
@@ -217,6 +226,9 @@ struct bcm_qspi {
 	struct bcm_qspi_dev_id *dev_ids;
 	struct completion mspi_done;
 	struct completion bspi_done;
+	u8 mspi_maj_rev;
+	u8 mspi_min_rev;
+	bool mspi_spcr3_sysclk;
 };
 
 static inline bool has_bspi(struct bcm_qspi *qspi)
@@ -224,6 +236,36 @@ static inline bool has_bspi(struct bcm_qspi *qspi)
 	return qspi->bspi_mode;
 }
 
+/* hardware supports spcr3 and fast baud-rate  */
+static inline bool bcm_qspi_has_fastbr(struct bcm_qspi *qspi)
+{
+	if (!has_bspi(qspi) &&
+	    ((qspi->mspi_maj_rev >= 1) &&
+	     (qspi->mspi_min_rev >= 5)))
+		return true;
+
+	return false;
+}
+
+/* hardware supports sys clk 108Mhz  */
+static inline bool bcm_qspi_has_sysclk_108(struct bcm_qspi *qspi)
+{
+	if (!has_bspi(qspi) && (qspi->mspi_spcr3_sysclk ||
+	    ((qspi->mspi_maj_rev >= 1) &&
+	     (qspi->mspi_min_rev >= 6))))
+		return true;
+
+	return false;
+}
+
+static inline int bcm_qspi_spbr_min(struct bcm_qspi *qspi)
+{
+	if (bcm_qspi_has_fastbr(qspi))
+		return 1;
+	else
+		return 8;
+}
+
 /* Read qspi controller register*/
 static inline u32 bcm_qspi_read(struct bcm_qspi *qspi, enum base_type type,
 				unsigned int offset)
@@ -531,16 +573,39 @@ static void bcm_qspi_hw_set_parms(struct bcm_qspi *qspi,
 	if (xp->speed_hz)
 		spbr = qspi->base_clk / (2 * xp->speed_hz);
 
-	spcr = clamp_val(spbr, QSPI_SPBR_MIN, QSPI_SPBR_MAX);
+	spcr = clamp_val(spbr, bcm_qspi_spbr_min(qspi), QSPI_SPBR_MAX);
 	bcm_qspi_write(qspi, MSPI, MSPI_SPCR0_LSB, spcr);
 
-	spcr = MSPI_MASTER_BIT;
+	if (!qspi->mspi_maj_rev)
+		/* legacy controller */
+		spcr = MSPI_MASTER_BIT;
+	else
+		spcr = 0;
+
 	/* for 16 bit the data should be zero */
 	if (xp->bits_per_word != 16)
 		spcr |= xp->bits_per_word << 2;
 	spcr |= xp->mode & 3;
+
 	bcm_qspi_write(qspi, MSPI, MSPI_SPCR0_MSB, spcr);
 
+	if (bcm_qspi_has_fastbr(qspi)) {
+		spcr = 0;
+
+		/* enable fastbr */
+		spcr |=	MSPI_SPCR3_FASTBR;
+
+		if (bcm_qspi_has_sysclk_108(qspi)) {
+			/* SYSCLK_108 */
+			spcr |= MSPI_SPCR3_SYSCLKSEL_108;
+			qspi->base_clk = MSPI_BASE_FREQ * 4;
+			/* Change spbr as we changed sysclk */
+			bcm_qspi_write(qspi, MSPI, MSPI_SPCR0_LSB, 4);
+		}
+
+		bcm_qspi_write(qspi, MSPI, MSPI_SPCR3, spcr);
+	}
+
 	qspi->last_parms = *xp;
 }
 
@@ -612,19 +677,15 @@ static int update_qspi_trans_byte_count(struct bcm_qspi *qspi,
 		if (qt->trans->cs_change &&
 		    (flags & TRANS_STATUS_BREAK_CS_CHANGE))
 			ret |= TRANS_STATUS_BREAK_CS_CHANGE;
-		if (ret)
-			goto done;
 
-		dev_dbg(&qspi->pdev->dev, "advance msg exit\n");
 		if (bcm_qspi_mspi_transfer_is_last(qspi, qt))
-			ret = TRANS_STATUS_BREAK_EOM;
+			ret |= TRANS_STATUS_BREAK_EOM;
 		else
-			ret = TRANS_STATUS_BREAK_NO_BYTES;
+			ret |= TRANS_STATUS_BREAK_NO_BYTES;
 
 		qt->trans = NULL;
 	}
 
-done:
 	dev_dbg(&qspi->pdev->dev, "trans %p len %d byte %d ret %x\n",
 		qt->trans, qt->trans ? qt->trans->len : 0, qt->byte, ret);
 	return ret;
@@ -670,7 +731,7 @@ static void read_from_hw(struct bcm_qspi *qspi, int slots)
 			if (buf)
 				buf[tp.byte] = read_rxram_slot_u8(qspi, slot);
 			dev_dbg(&qspi->pdev->dev, "RD %02x\n",
-				buf ? buf[tp.byte] : 0xff);
+				buf ? buf[tp.byte] : 0x0);
 		} else {
 			u16 *buf = tp.trans->rx_buf;
 
@@ -678,7 +739,7 @@ static void read_from_hw(struct bcm_qspi *qspi, int slots)
 				buf[tp.byte / 2] = read_rxram_slot_u16(qspi,
 								      slot);
 			dev_dbg(&qspi->pdev->dev, "RD %04x\n",
-				buf ? buf[tp.byte] : 0xffff);
+				buf ? buf[tp.byte / 2] : 0x0);
 		}
 
 		update_qspi_trans_byte_count(qspi, &tp,
@@ -733,13 +794,13 @@ static int write_to_hw(struct bcm_qspi *qspi, struct spi_device *spi)
 	while (!tstatus && slot < MSPI_NUM_CDRAM) {
 		if (tp.trans->bits_per_word <= 8) {
 			const u8 *buf = tp.trans->tx_buf;
-			u8 val = buf ? buf[tp.byte] : 0xff;
+			u8 val = buf ? buf[tp.byte] : 0x00;
 
 			write_txram_slot_u8(qspi, slot, val);
 			dev_dbg(&qspi->pdev->dev, "WR %02x\n", val);
 		} else {
 			const u16 *buf = tp.trans->tx_buf;
-			u16 val = buf ? buf[tp.byte / 2] : 0xffff;
+			u16 val = buf ? buf[tp.byte / 2] : 0x0000;
 
 			write_txram_slot_u16(qspi, slot, val);
 			dev_dbg(&qspi->pdev->dev, "WR %04x\n", val);
@@ -771,7 +832,16 @@ static int write_to_hw(struct bcm_qspi *qspi, struct spi_device *spi)
 	bcm_qspi_write(qspi, MSPI, MSPI_NEWQP, 0);
 	bcm_qspi_write(qspi, MSPI, MSPI_ENDQP, slot - 1);
 
-	if (tstatus & TRANS_STATUS_BREAK_DESELECT) {
+	/*
+	 *  case 1) EOM =1, cs_change =0: SSb inactive
+	 *  case 2) EOM =1, cs_change =1: SSb stay active
+	 *  case 3) EOM =0, cs_change =0: SSb stay active
+	 *  case 4) EOM =0, cs_change =1: SSb inactive
+	 */
+	if (((tstatus & TRANS_STATUS_BREAK_DESELECT)
+	     == TRANS_STATUS_BREAK_CS_CHANGE) ||
+	    ((tstatus & TRANS_STATUS_BREAK_DESELECT)
+	     == TRANS_STATUS_BREAK_EOM)) {
 		mspi_cdram = read_cdram_slot(qspi, slot - 1) &
 			~MSPI_CDRAM_CONT_BIT;
 		write_cdram_slot(qspi, slot - 1, mspi_cdram);
@@ -1190,8 +1260,51 @@ static const struct spi_controller_mem_ops bcm_qspi_mem_ops = {
 	.exec_op = bcm_qspi_exec_mem_op,
 };
 
+struct bcm_qspi_data {
+	bool	has_mspi_rev;
+	bool	has_spcr3_sysclk;
+};
+
+static const struct bcm_qspi_data bcm_qspi_no_rev_data = {
+	.has_mspi_rev	= false,
+	.has_spcr3_sysclk = false,
+};
+
+static const struct bcm_qspi_data bcm_qspi_rev_data = {
+	.has_mspi_rev	= true,
+	.has_spcr3_sysclk = false,
+};
+
+static const struct bcm_qspi_data bcm_qspi_spcr3_data = {
+	.has_mspi_rev	= true,
+	.has_spcr3_sysclk = true,
+};
+
 static const struct of_device_id bcm_qspi_of_match[] = {
-	{ .compatible = "brcm,spi-bcm-qspi" },
+	{
+		.compatible = "brcm,spi-bcm7425-qspi",
+		.data = &bcm_qspi_no_rev_data,
+	},
+	{
+		.compatible = "brcm,spi-bcm7429-qspi",
+		.data = &bcm_qspi_no_rev_data,
+	},
+	{
+		.compatible = "brcm,spi-bcm7435-qspi",
+		.data = &bcm_qspi_no_rev_data,
+	},
+	{
+		.compatible = "brcm,spi-bcm-qspi",
+		.data = &bcm_qspi_rev_data,
+	},
+	{
+		.compatible = "brcm,spi-bcm7216-qspi",
+		.data = &bcm_qspi_spcr3_data,
+	},
+	{
+		.compatible = "brcm,spi-bcm7278-qspi",
+		.data = &bcm_qspi_spcr3_data,
+	},
 	{},
 };
 MODULE_DEVICE_TABLE(of, bcm_qspi_of_match);
@@ -1199,12 +1312,15 @@ MODULE_DEVICE_TABLE(of, bcm_qspi_of_match);
 int bcm_qspi_probe(struct platform_device *pdev,
 		   struct bcm_qspi_soc_intc *soc_intc)
 {
+	const struct of_device_id *of_id = NULL;
+	const struct bcm_qspi_data *data;
 	struct device *dev = &pdev->dev;
 	struct bcm_qspi *qspi;
 	struct spi_master *master;
 	struct resource *res;
 	int irq, ret = 0, num_ints = 0;
 	u32 val;
+	u32 rev = 0;
 	const char *name = NULL;
 	int num_irqs = ARRAY_SIZE(qspi_irq_tab);
 
@@ -1212,9 +1328,12 @@ int bcm_qspi_probe(struct platform_device *pdev,
 	if (!dev->of_node)
 		return -ENODEV;
 
-	if (!of_match_node(bcm_qspi_of_match, dev->of_node))
+	of_id = of_match_node(bcm_qspi_of_match, dev->of_node);
+	if (!of_id)
 		return -ENODEV;
 
+	data = of_id->data;
+
 	master = spi_alloc_master(dev, sizeof(struct bcm_qspi));
 	if (!master) {
 		dev_err(dev, "error allocating spi_master\n");
@@ -1222,6 +1341,11 @@ int bcm_qspi_probe(struct platform_device *pdev,
 	}
 
 	qspi = spi_master_get_devdata(master);
+
+	qspi->clk = devm_clk_get_optional(&pdev->dev, NULL);
+	if (IS_ERR(qspi->clk))
+		return PTR_ERR(qspi->clk);
+
 	qspi->pdev = pdev;
 	qspi->trans_pos.trans = NULL;
 	qspi->trans_pos.byte = 0;
@@ -1335,13 +1459,6 @@ int bcm_qspi_probe(struct platform_device *pdev,
 		qspi->soc_intc = NULL;
 	}
 
-	qspi->clk = devm_clk_get(&pdev->dev, NULL);
-	if (IS_ERR(qspi->clk)) {
-		dev_warn(dev, "unable to get clock\n");
-		ret = PTR_ERR(qspi->clk);
-		goto qspi_probe_err;
-	}
-
 	ret = clk_prepare_enable(qspi->clk);
 	if (ret) {
 		dev_err(dev, "failed to prepare clock\n");
@@ -1349,7 +1466,19 @@ int bcm_qspi_probe(struct platform_device *pdev,
 	}
 
 	qspi->base_clk = clk_get_rate(qspi->clk);
-	qspi->max_speed_hz = qspi->base_clk / (QSPI_SPBR_MIN * 2);
+
+	if (data->has_mspi_rev) {
+		rev = bcm_qspi_read(qspi, MSPI, MSPI_REV);
+		/* some older revs do not have a MSPI_REV register */
+		if ((rev & 0xff) == 0xff)
+			rev = 0;
+	}
+
+	qspi->mspi_maj_rev = (rev >> 4) & 0xf;
+	qspi->mspi_min_rev = rev & 0xf;
+	qspi->mspi_spcr3_sysclk = data->has_spcr3_sysclk;
+
+	qspi->max_speed_hz = qspi->base_clk / (bcm_qspi_spbr_min(qspi) * 2);
 
 	bcm_qspi_hw_init(qspi);
 	init_completion(&qspi->mspi_done);
@@ -1406,7 +1535,7 @@ static int __maybe_unused bcm_qspi_suspend(struct device *dev)
 			bcm_qspi_read(qspi, BSPI, BSPI_STRAP_OVERRIDE_CTRL);
 
 	spi_master_suspend(qspi->master);
-	clk_disable(qspi->clk);
+	clk_disable_unprepare(qspi->clk);
 	bcm_qspi_hw_uninit(qspi);
 
 	return 0;
@@ -1424,7 +1553,7 @@ static int __maybe_unused bcm_qspi_resume(struct device *dev)
 		qspi->soc_intc->bcm_qspi_int_set(qspi->soc_intc, MSPI_DONE,
 						 true);
 
-	ret = clk_enable(qspi->clk);
+	ret = clk_prepare_enable(qspi->clk);
 	if (!ret)
 		spi_master_resume(qspi->master);
 
diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c
index 11c235879bb7..237bd306c268 100644
--- a/drivers/spi/spi-bcm2835.c
+++ b/drivers/spi/spi-bcm2835.c
@@ -191,12 +191,12 @@ static void bcm2835_debugfs_remove(struct bcm2835_spi *bs)
 }
 #endif /* CONFIG_DEBUG_FS */
 
-static inline u32 bcm2835_rd(struct bcm2835_spi *bs, unsigned reg)
+static inline u32 bcm2835_rd(struct bcm2835_spi *bs, unsigned int reg)
 {
 	return readl(bs->regs + reg);
 }
 
-static inline void bcm2835_wr(struct bcm2835_spi *bs, unsigned reg, u32 val)
+static inline void bcm2835_wr(struct bcm2835_spi *bs, unsigned int reg, u32 val)
 {
 	writel(val, bs->regs + reg);
 }
@@ -940,6 +940,7 @@ static int bcm2835_dma_init(struct spi_controller *ctlr, struct device *dev,
 	if (dma_mapping_error(ctlr->dma_tx->device->dev, bs->fill_tx_addr)) {
 		dev_err(dev, "cannot map zero page - not using DMA mode\n");
 		bs->fill_tx_addr = 0;
+		ret = -ENOMEM;
 		goto err_release;
 	}
 
@@ -949,6 +950,7 @@ static int bcm2835_dma_init(struct spi_controller *ctlr, struct device *dev,
 						     DMA_MEM_TO_DEV, 0);
 	if (!bs->fill_tx_desc) {
 		dev_err(dev, "cannot prepare fill_tx_desc - not using DMA mode\n");
+		ret = -ENOMEM;
 		goto err_release;
 	}
 
@@ -979,6 +981,7 @@ static int bcm2835_dma_init(struct spi_controller *ctlr, struct device *dev,
 	if (dma_mapping_error(ctlr->dma_rx->device->dev, bs->clear_rx_addr)) {
 		dev_err(dev, "cannot map clear_rx_cs - not using DMA mode\n");
 		bs->clear_rx_addr = 0;
+		ret = -ENOMEM;
 		goto err_release;
 	}
 
@@ -989,6 +992,7 @@ static int bcm2835_dma_init(struct spi_controller *ctlr, struct device *dev,
 					   DMA_MEM_TO_DEV, 0);
 		if (!bs->clear_rx_desc[i]) {
 			dev_err(dev, "cannot prepare clear_rx_desc - not using DMA mode\n");
+			ret = -ENOMEM;
 			goto err_release;
 		}
 
@@ -1347,7 +1351,7 @@ static int bcm2835_spi_probe(struct platform_device *pdev)
 		goto out_dma_release;
 	}
 
-	err = devm_spi_register_controller(&pdev->dev, ctlr);
+	err = spi_register_controller(ctlr);
 	if (err) {
 		dev_err(&pdev->dev, "could not register SPI controller: %d\n",
 			err);
@@ -1374,17 +1378,28 @@ static int bcm2835_spi_remove(struct platform_device *pdev)
 
 	bcm2835_debugfs_remove(bs);
 
+	spi_unregister_controller(ctlr);
+
+	bcm2835_dma_release(ctlr, bs);
+
 	/* Clear FIFOs, and disable the HW block */
 	bcm2835_wr(bs, BCM2835_SPI_CS,
 		   BCM2835_SPI_CS_CLEAR_RX | BCM2835_SPI_CS_CLEAR_TX);
 
 	clk_disable_unprepare(bs->clk);
 
-	bcm2835_dma_release(ctlr, bs);
-
 	return 0;
 }
 
+static void bcm2835_spi_shutdown(struct platform_device *pdev)
+{
+	int ret;
+
+	ret = bcm2835_spi_remove(pdev);
+	if (ret)
+		dev_err(&pdev->dev, "failed to shutdown\n");
+}
+
 static const struct of_device_id bcm2835_spi_match[] = {
 	{ .compatible = "brcm,bcm2835-spi", },
 	{}
@@ -1398,6 +1413,7 @@ static struct platform_driver bcm2835_spi_driver = {
 	},
 	.probe		= bcm2835_spi_probe,
 	.remove		= bcm2835_spi_remove,
+	.shutdown	= bcm2835_spi_shutdown,
 };
 module_platform_driver(bcm2835_spi_driver);
 
diff --git a/drivers/spi/spi-bcm2835aux.c b/drivers/spi/spi-bcm2835aux.c
index a2162ff56a12..c331efd6e86b 100644
--- a/drivers/spi/spi-bcm2835aux.c
+++ b/drivers/spi/spi-bcm2835aux.c
@@ -569,7 +569,7 @@ static int bcm2835aux_spi_probe(struct platform_device *pdev)
 		goto out_clk_disable;
 	}
 
-	err = devm_spi_register_master(&pdev->dev, master);
+	err = spi_register_master(master);
 	if (err) {
 		dev_err(&pdev->dev, "could not register SPI master: %d\n", err);
 		goto out_clk_disable;
@@ -593,6 +593,8 @@ static int bcm2835aux_spi_remove(struct platform_device *pdev)
 
 	bcm2835aux_debugfs_remove(bs);
 
+	spi_unregister_master(master);
+
 	bcm2835aux_spi_reset_hw(bs);
 
 	/* disable the HW block by releasing the clock */
diff --git a/drivers/spi/spi-dw.c b/drivers/spi/spi-dw-core.c
index 31e3f866d11a..323c66c5db50 100644
--- a/drivers/spi/spi-dw.c
+++ b/drivers/spi/spi-dw-core.c
@@ -24,74 +24,34 @@ struct chip_data {
 	u8 tmode;		/* TR/TO/RO/EEPROM */
 	u8 type;		/* SPI/SSP/MicroWire */
 
-	u8 poll_mode;		/* 1 means use poll mode */
-
 	u16 clk_div;		/* baud rate divider */
 	u32 speed_hz;		/* baud rate */
-	void (*cs_control)(u32 command);
 };
 
 #ifdef CONFIG_DEBUG_FS
-#define SPI_REGS_BUFSIZE	1024
-static ssize_t dw_spi_show_regs(struct file *file, char __user *user_buf,
-		size_t count, loff_t *ppos)
-{
-	struct dw_spi *dws = file->private_data;
-	char *buf;
-	u32 len = 0;
-	ssize_t ret;
-
-	buf = kzalloc(SPI_REGS_BUFSIZE, GFP_KERNEL);
-	if (!buf)
-		return 0;
-
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"%s registers:\n", dev_name(&dws->master->dev));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"=================================\n");
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"CTRL0: \t\t0x%08x\n", dw_readl(dws, DW_SPI_CTRL0));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"CTRL1: \t\t0x%08x\n", dw_readl(dws, DW_SPI_CTRL1));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"SSIENR: \t0x%08x\n", dw_readl(dws, DW_SPI_SSIENR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"SER: \t\t0x%08x\n", dw_readl(dws, DW_SPI_SER));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"BAUDR: \t\t0x%08x\n", dw_readl(dws, DW_SPI_BAUDR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"TXFTLR: \t0x%08x\n", dw_readl(dws, DW_SPI_TXFLTR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"RXFTLR: \t0x%08x\n", dw_readl(dws, DW_SPI_RXFLTR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"TXFLR: \t\t0x%08x\n", dw_readl(dws, DW_SPI_TXFLR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"RXFLR: \t\t0x%08x\n", dw_readl(dws, DW_SPI_RXFLR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"SR: \t\t0x%08x\n", dw_readl(dws, DW_SPI_SR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"IMR: \t\t0x%08x\n", dw_readl(dws, DW_SPI_IMR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"ISR: \t\t0x%08x\n", dw_readl(dws, DW_SPI_ISR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"DMACR: \t\t0x%08x\n", dw_readl(dws, DW_SPI_DMACR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"DMATDLR: \t0x%08x\n", dw_readl(dws, DW_SPI_DMATDLR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"DMARDLR: \t0x%08x\n", dw_readl(dws, DW_SPI_DMARDLR));
-	len += scnprintf(buf + len, SPI_REGS_BUFSIZE - len,
-			"=================================\n");
-
-	ret = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-	return ret;
+
+#define DW_SPI_DBGFS_REG(_name, _off)	\
+{					\
+	.name = _name,			\
+	.offset = _off,			\
 }
 
-static const struct file_operations dw_spi_regs_ops = {
-	.owner		= THIS_MODULE,
-	.open		= simple_open,
-	.read		= dw_spi_show_regs,
-	.llseek		= default_llseek,
+static const struct debugfs_reg32 dw_spi_dbgfs_regs[] = {
+	DW_SPI_DBGFS_REG("CTRLR0", DW_SPI_CTRLR0),
+	DW_SPI_DBGFS_REG("CTRLR1", DW_SPI_CTRLR1),
+	DW_SPI_DBGFS_REG("SSIENR", DW_SPI_SSIENR),
+	DW_SPI_DBGFS_REG("SER", DW_SPI_SER),
+	DW_SPI_DBGFS_REG("BAUDR", DW_SPI_BAUDR),
+	DW_SPI_DBGFS_REG("TXFTLR", DW_SPI_TXFTLR),
+	DW_SPI_DBGFS_REG("RXFTLR", DW_SPI_RXFTLR),
+	DW_SPI_DBGFS_REG("TXFLR", DW_SPI_TXFLR),
+	DW_SPI_DBGFS_REG("RXFLR", DW_SPI_RXFLR),
+	DW_SPI_DBGFS_REG("SR", DW_SPI_SR),
+	DW_SPI_DBGFS_REG("IMR", DW_SPI_IMR),
+	DW_SPI_DBGFS_REG("ISR", DW_SPI_ISR),
+	DW_SPI_DBGFS_REG("DMACR", DW_SPI_DMACR),
+	DW_SPI_DBGFS_REG("DMATDLR", DW_SPI_DMATDLR),
+	DW_SPI_DBGFS_REG("DMARDLR", DW_SPI_DMARDLR),
 };
 
 static int dw_spi_debugfs_init(struct dw_spi *dws)
@@ -103,8 +63,11 @@ static int dw_spi_debugfs_init(struct dw_spi *dws)
 	if (!dws->debugfs)
 		return -ENOMEM;
 
-	debugfs_create_file("registers", S_IFREG | S_IRUGO,
-		dws->debugfs, (void *)dws, &dw_spi_regs_ops);
+	dws->regset.regs = dw_spi_dbgfs_regs;
+	dws->regset.nregs = ARRAY_SIZE(dw_spi_dbgfs_regs);
+	dws->regset.base = dws->regs;
+	debugfs_create_regset32("registers", 0400, dws->debugfs, &dws->regset);
+
 	return 0;
 }
 
@@ -127,13 +90,16 @@ static inline void dw_spi_debugfs_remove(struct dw_spi *dws)
 void dw_spi_set_cs(struct spi_device *spi, bool enable)
 {
 	struct dw_spi *dws = spi_controller_get_devdata(spi->controller);
-	struct chip_data *chip = spi_get_ctldata(spi);
-
-	/* Chip select logic is inverted from spi_set_cs() */
-	if (chip && chip->cs_control)
-		chip->cs_control(!enable);
+	bool cs_high = !!(spi->mode & SPI_CS_HIGH);
 
-	if (!enable)
+	/*
+	 * DW SPI controller demands any native CS being set in order to
+	 * proceed with data transfer. So in order to activate the SPI
+	 * communications we must set a corresponding bit in the Slave
+	 * Enable register no matter whether the SPI core is configured to
+	 * support active-high or active-low CS level.
+	 */
+	if (cs_high == enable)
 		dw_writel(dws, DW_SPI_SER, BIT(spi->chip_select));
 	else if (dws->cs_override)
 		dw_writel(dws, DW_SPI_SER, 0);
@@ -265,17 +231,56 @@ static irqreturn_t dw_spi_irq(int irq, void *dev_id)
 	return dws->transfer_handler(dws);
 }
 
-/* Must be called inside pump_transfers() */
-static int poll_transfer(struct dw_spi *dws)
+/* Configure CTRLR0 for DW_apb_ssi */
+u32 dw_spi_update_cr0(struct spi_controller *master, struct spi_device *spi,
+		      struct spi_transfer *transfer)
 {
-	do {
-		dw_writer(dws);
-		dw_reader(dws);
-		cpu_relax();
-	} while (dws->rx_end > dws->rx);
+	struct chip_data *chip = spi_get_ctldata(spi);
+	u32 cr0;
 
-	return 0;
+	/* Default SPI mode is SCPOL = 0, SCPH = 0 */
+	cr0 = (transfer->bits_per_word - 1)
+		| (chip->type << SPI_FRF_OFFSET)
+		| ((((spi->mode & SPI_CPOL) ? 1 : 0) << SPI_SCOL_OFFSET) |
+		   (((spi->mode & SPI_CPHA) ? 1 : 0) << SPI_SCPH_OFFSET) |
+		   (((spi->mode & SPI_LOOP) ? 1 : 0) << SPI_SRL_OFFSET))
+		| (chip->tmode << SPI_TMOD_OFFSET);
+
+	return cr0;
+}
+EXPORT_SYMBOL_GPL(dw_spi_update_cr0);
+
+/* Configure CTRLR0 for DWC_ssi */
+u32 dw_spi_update_cr0_v1_01a(struct spi_controller *master,
+			     struct spi_device *spi,
+			     struct spi_transfer *transfer)
+{
+	struct chip_data *chip = spi_get_ctldata(spi);
+	u32 cr0;
+
+	/* CTRLR0[ 4: 0] Data Frame Size */
+	cr0 = (transfer->bits_per_word - 1);
+
+	/* CTRLR0[ 7: 6] Frame Format */
+	cr0 |= chip->type << DWC_SSI_CTRLR0_FRF_OFFSET;
+
+	/*
+	 * SPI mode (SCPOL|SCPH)
+	 * CTRLR0[ 8] Serial Clock Phase
+	 * CTRLR0[ 9] Serial Clock Polarity
+	 */
+	cr0 |= ((spi->mode & SPI_CPOL) ? 1 : 0) << DWC_SSI_CTRLR0_SCPOL_OFFSET;
+	cr0 |= ((spi->mode & SPI_CPHA) ? 1 : 0) << DWC_SSI_CTRLR0_SCPH_OFFSET;
+
+	/* CTRLR0[11:10] Transfer Mode */
+	cr0 |= chip->tmode << DWC_SSI_CTRLR0_TMOD_OFFSET;
+
+	/* CTRLR0[13] Shift Register Loop */
+	cr0 |= ((spi->mode & SPI_LOOP) ? 1 : 0) << DWC_SSI_CTRLR0_SRL_OFFSET;
+
+	return cr0;
 }
+EXPORT_SYMBOL_GPL(dw_spi_update_cr0_v1_01a);
 
 static int dw_spi_transfer_one(struct spi_controller *master,
 		struct spi_device *spi, struct spi_transfer *transfer)
@@ -313,34 +318,11 @@ static int dw_spi_transfer_one(struct spi_controller *master,
 		spi_set_clk(dws, chip->clk_div);
 	}
 
+	transfer->effective_speed_hz = dws->max_freq / chip->clk_div;
 	dws->n_bytes = DIV_ROUND_UP(transfer->bits_per_word, BITS_PER_BYTE);
-	dws->dma_width = DIV_ROUND_UP(transfer->bits_per_word, BITS_PER_BYTE);
-
-	/* Default SPI mode is SCPOL = 0, SCPH = 0 */
-	cr0 = (transfer->bits_per_word - 1)
-		| (chip->type << SPI_FRF_OFFSET)
-		| ((((spi->mode & SPI_CPOL) ? 1 : 0) << SPI_SCOL_OFFSET) |
-			(((spi->mode & SPI_CPHA) ? 1 : 0) << SPI_SCPH_OFFSET) |
-			(((spi->mode & SPI_LOOP) ? 1 : 0) << SPI_SRL_OFFSET))
-		| (chip->tmode << SPI_TMOD_OFFSET);
 
-	/*
-	 * Adjust transfer mode if necessary. Requires platform dependent
-	 * chipselect mechanism.
-	 */
-	if (chip->cs_control) {
-		if (dws->rx && dws->tx)
-			chip->tmode = SPI_TMOD_TR;
-		else if (dws->rx)
-			chip->tmode = SPI_TMOD_RO;
-		else
-			chip->tmode = SPI_TMOD_TO;
-
-		cr0 &= ~SPI_TMOD_MASK;
-		cr0 |= (chip->tmode << SPI_TMOD_OFFSET);
-	}
-
-	dw_writel(dws, DW_SPI_CTRL0, cr0);
+	cr0 = dws->update_cr0(master, spi, transfer);
+	dw_writel(dws, DW_SPI_CTRLR0, cr0);
 
 	/* Check if current transfer is a DMA transaction */
 	if (master->can_dma && master->can_dma(master, spi, transfer))
@@ -359,9 +341,9 @@ static int dw_spi_transfer_one(struct spi_controller *master,
 			spi_enable_chip(dws, 1);
 			return ret;
 		}
-	} else if (!chip->poll_mode) {
+	} else {
 		txlevel = min_t(u16, dws->fifo_len / 2, dws->len / dws->n_bytes);
-		dw_writel(dws, DW_SPI_TXFLTR, txlevel);
+		dw_writel(dws, DW_SPI_TXFTLR, txlevel);
 
 		/* Set the interrupt mask */
 		imask |= SPI_INT_TXEI | SPI_INT_TXOI |
@@ -373,14 +355,8 @@ static int dw_spi_transfer_one(struct spi_controller *master,
 
 	spi_enable_chip(dws, 1);
 
-	if (dws->dma_mapped) {
-		ret = dws->dma_ops->dma_transfer(dws, transfer);
-		if (ret < 0)
-			return ret;
-	}
-
-	if (chip->poll_mode)
-		return poll_transfer(dws);
+	if (dws->dma_mapped)
+		return dws->dma_ops->dma_transfer(dws, transfer);
 
 	return 1;
 }
@@ -399,7 +375,6 @@ static void dw_spi_handle_err(struct spi_controller *master,
 /* This may be called twice for each spi dev */
 static int dw_spi_setup(struct spi_device *spi)
 {
-	struct dw_spi_chip *chip_info = NULL;
 	struct chip_data *chip;
 
 	/* Only alloc on first setup */
@@ -411,21 +386,6 @@ static int dw_spi_setup(struct spi_device *spi)
 		spi_set_ctldata(spi, chip);
 	}
 
-	/*
-	 * Protocol drivers may change the chip settings, so...
-	 * if chip_info exists, use it
-	 */
-	chip_info = spi->controller_data;
-
-	/* chip_info doesn't always exist */
-	if (chip_info) {
-		if (chip_info->cs_control)
-			chip->cs_control = chip_info->cs_control;
-
-		chip->poll_mode = chip_info->poll_mode;
-		chip->type = chip_info->type;
-	}
-
 	chip->tmode = SPI_TMOD_TR;
 
 	return 0;
@@ -452,11 +412,11 @@ static void spi_hw_init(struct device *dev, struct dw_spi *dws)
 		u32 fifo;
 
 		for (fifo = 1; fifo < 256; fifo++) {
-			dw_writel(dws, DW_SPI_TXFLTR, fifo);
-			if (fifo != dw_readl(dws, DW_SPI_TXFLTR))
+			dw_writel(dws, DW_SPI_TXFTLR, fifo);
+			if (fifo != dw_readl(dws, DW_SPI_TXFTLR))
 				break;
 		}
-		dw_writel(dws, DW_SPI_TXFLTR, 0);
+		dw_writel(dws, DW_SPI_TXFTLR, 0);
 
 		dws->fifo_len = (fifo == 1) ? 0 : fifo;
 		dev_dbg(dev, "Detected FIFO size: %u bytes\n", dws->fifo_len);
@@ -481,7 +441,6 @@ int dw_spi_add_host(struct device *dev, struct dw_spi *dws)
 
 	dws->master = master;
 	dws->type = SSI_MOTO_SPI;
-	dws->dma_inited = 0;
 	dws->dma_addr = (dma_addr_t)(dws->paddr + DW_SPI_DR);
 	spin_lock_init(&dws->buf_lock);
 
@@ -517,16 +476,16 @@ int dw_spi_add_host(struct device *dev, struct dw_spi *dws)
 	spi_hw_init(dev, dws);
 
 	if (dws->dma_ops && dws->dma_ops->dma_init) {
-		ret = dws->dma_ops->dma_init(dws);
+		ret = dws->dma_ops->dma_init(dev, dws);
 		if (ret) {
 			dev_warn(dev, "DMA init failed\n");
-			dws->dma_inited = 0;
 		} else {
 			master->can_dma = dws->dma_ops->can_dma;
+			master->flags |= SPI_CONTROLLER_MUST_TX;
 		}
 	}
 
-	ret = devm_spi_register_controller(dev, master);
+	ret = spi_register_controller(master);
 	if (ret) {
 		dev_err(&master->dev, "problem registering spi master\n");
 		goto err_dma_exit;
@@ -550,6 +509,8 @@ void dw_spi_remove_host(struct dw_spi *dws)
 {
 	dw_spi_debugfs_remove(dws);
 
+	spi_unregister_controller(dws->master);
+
 	if (dws->dma_ops && dws->dma_ops->dma_exit)
 		dws->dma_ops->dma_exit(dws);
 
diff --git a/drivers/spi/spi-dw-dma.c b/drivers/spi/spi-dw-dma.c
new file mode 100644
index 000000000000..5986c520b196
--- /dev/null
+++ b/drivers/spi/spi-dw-dma.c
@@ -0,0 +1,480 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Special handling for DW DMA core
+ *
+ * Copyright (c) 2009, 2014 Intel Corporation.
+ */
+
+#include <linux/completion.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmaengine.h>
+#include <linux/irqreturn.h>
+#include <linux/jiffies.h>
+#include <linux/pci.h>
+#include <linux/platform_data/dma-dw.h>
+#include <linux/spi/spi.h>
+#include <linux/types.h>
+
+#include "spi-dw.h"
+
+#define WAIT_RETRIES	5
+#define RX_BUSY		0
+#define RX_BURST_LEVEL	16
+#define TX_BUSY		1
+#define TX_BURST_LEVEL	16
+
+static bool dw_spi_dma_chan_filter(struct dma_chan *chan, void *param)
+{
+	struct dw_dma_slave *s = param;
+
+	if (s->dma_dev != chan->device->dev)
+		return false;
+
+	chan->private = s;
+	return true;
+}
+
+static void dw_spi_dma_maxburst_init(struct dw_spi *dws)
+{
+	struct dma_slave_caps caps;
+	u32 max_burst, def_burst;
+	int ret;
+
+	def_burst = dws->fifo_len / 2;
+
+	ret = dma_get_slave_caps(dws->rxchan, &caps);
+	if (!ret && caps.max_burst)
+		max_burst = caps.max_burst;
+	else
+		max_burst = RX_BURST_LEVEL;
+
+	dws->rxburst = min(max_burst, def_burst);
+
+	ret = dma_get_slave_caps(dws->txchan, &caps);
+	if (!ret && caps.max_burst)
+		max_burst = caps.max_burst;
+	else
+		max_burst = TX_BURST_LEVEL;
+
+	dws->txburst = min(max_burst, def_burst);
+}
+
+static int dw_spi_dma_init_mfld(struct device *dev, struct dw_spi *dws)
+{
+	struct dw_dma_slave dma_tx = { .dst_id = 1 }, *tx = &dma_tx;
+	struct dw_dma_slave dma_rx = { .src_id = 0 }, *rx = &dma_rx;
+	struct pci_dev *dma_dev;
+	dma_cap_mask_t mask;
+
+	/*
+	 * Get pci device for DMA controller, currently it could only
+	 * be the DMA controller of Medfield
+	 */
+	dma_dev = pci_get_device(PCI_VENDOR_ID_INTEL, 0x0827, NULL);
+	if (!dma_dev)
+		return -ENODEV;
+
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+
+	/* 1. Init rx channel */
+	rx->dma_dev = &dma_dev->dev;
+	dws->rxchan = dma_request_channel(mask, dw_spi_dma_chan_filter, rx);
+	if (!dws->rxchan)
+		goto err_exit;
+
+	/* 2. Init tx channel */
+	tx->dma_dev = &dma_dev->dev;
+	dws->txchan = dma_request_channel(mask, dw_spi_dma_chan_filter, tx);
+	if (!dws->txchan)
+		goto free_rxchan;
+
+	dws->master->dma_rx = dws->rxchan;
+	dws->master->dma_tx = dws->txchan;
+
+	init_completion(&dws->dma_completion);
+
+	dw_spi_dma_maxburst_init(dws);
+
+	return 0;
+
+free_rxchan:
+	dma_release_channel(dws->rxchan);
+	dws->rxchan = NULL;
+err_exit:
+	return -EBUSY;
+}
+
+static int dw_spi_dma_init_generic(struct device *dev, struct dw_spi *dws)
+{
+	dws->rxchan = dma_request_slave_channel(dev, "rx");
+	if (!dws->rxchan)
+		return -ENODEV;
+
+	dws->txchan = dma_request_slave_channel(dev, "tx");
+	if (!dws->txchan) {
+		dma_release_channel(dws->rxchan);
+		dws->rxchan = NULL;
+		return -ENODEV;
+	}
+
+	dws->master->dma_rx = dws->rxchan;
+	dws->master->dma_tx = dws->txchan;
+
+	init_completion(&dws->dma_completion);
+
+	dw_spi_dma_maxburst_init(dws);
+
+	return 0;
+}
+
+static void dw_spi_dma_exit(struct dw_spi *dws)
+{
+	if (dws->txchan) {
+		dmaengine_terminate_sync(dws->txchan);
+		dma_release_channel(dws->txchan);
+	}
+
+	if (dws->rxchan) {
+		dmaengine_terminate_sync(dws->rxchan);
+		dma_release_channel(dws->rxchan);
+	}
+
+	dw_writel(dws, DW_SPI_DMACR, 0);
+}
+
+static irqreturn_t dw_spi_dma_transfer_handler(struct dw_spi *dws)
+{
+	u16 irq_status = dw_readl(dws, DW_SPI_ISR);
+
+	if (!irq_status)
+		return IRQ_NONE;
+
+	dw_readl(dws, DW_SPI_ICR);
+	spi_reset_chip(dws);
+
+	dev_err(&dws->master->dev, "%s: FIFO overrun/underrun\n", __func__);
+	dws->master->cur_msg->status = -EIO;
+	complete(&dws->dma_completion);
+	return IRQ_HANDLED;
+}
+
+static bool dw_spi_can_dma(struct spi_controller *master,
+			   struct spi_device *spi, struct spi_transfer *xfer)
+{
+	struct dw_spi *dws = spi_controller_get_devdata(master);
+
+	return xfer->len > dws->fifo_len;
+}
+
+static enum dma_slave_buswidth dw_spi_dma_convert_width(u8 n_bytes)
+{
+	if (n_bytes == 1)
+		return DMA_SLAVE_BUSWIDTH_1_BYTE;
+	else if (n_bytes == 2)
+		return DMA_SLAVE_BUSWIDTH_2_BYTES;
+
+	return DMA_SLAVE_BUSWIDTH_UNDEFINED;
+}
+
+static int dw_spi_dma_wait(struct dw_spi *dws, struct spi_transfer *xfer)
+{
+	unsigned long long ms;
+
+	ms = xfer->len * MSEC_PER_SEC * BITS_PER_BYTE;
+	do_div(ms, xfer->effective_speed_hz);
+	ms += ms + 200;
+
+	if (ms > UINT_MAX)
+		ms = UINT_MAX;
+
+	ms = wait_for_completion_timeout(&dws->dma_completion,
+					 msecs_to_jiffies(ms));
+
+	if (ms == 0) {
+		dev_err(&dws->master->cur_msg->spi->dev,
+			"DMA transaction timed out\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+static inline bool dw_spi_dma_tx_busy(struct dw_spi *dws)
+{
+	return !(dw_readl(dws, DW_SPI_SR) & SR_TF_EMPT);
+}
+
+static int dw_spi_dma_wait_tx_done(struct dw_spi *dws,
+				   struct spi_transfer *xfer)
+{
+	int retry = WAIT_RETRIES;
+	struct spi_delay delay;
+	u32 nents;
+
+	nents = dw_readl(dws, DW_SPI_TXFLR);
+	delay.unit = SPI_DELAY_UNIT_SCK;
+	delay.value = nents * dws->n_bytes * BITS_PER_BYTE;
+
+	while (dw_spi_dma_tx_busy(dws) && retry--)
+		spi_delay_exec(&delay, xfer);
+
+	if (retry < 0) {
+		dev_err(&dws->master->dev, "Tx hanged up\n");
+		return -EIO;
+	}
+
+	return 0;
+}
+
+/*
+ * dws->dma_chan_busy is set before the dma transfer starts, callback for tx
+ * channel will clear a corresponding bit.
+ */
+static void dw_spi_dma_tx_done(void *arg)
+{
+	struct dw_spi *dws = arg;
+
+	clear_bit(TX_BUSY, &dws->dma_chan_busy);
+	if (test_bit(RX_BUSY, &dws->dma_chan_busy))
+		return;
+
+	dw_writel(dws, DW_SPI_DMACR, 0);
+	complete(&dws->dma_completion);
+}
+
+static struct dma_async_tx_descriptor *
+dw_spi_dma_prepare_tx(struct dw_spi *dws, struct spi_transfer *xfer)
+{
+	struct dma_slave_config txconf;
+	struct dma_async_tx_descriptor *txdesc;
+
+	if (!xfer->tx_buf)
+		return NULL;
+
+	memset(&txconf, 0, sizeof(txconf));
+	txconf.direction = DMA_MEM_TO_DEV;
+	txconf.dst_addr = dws->dma_addr;
+	txconf.dst_maxburst = dws->txburst;
+	txconf.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	txconf.dst_addr_width = dw_spi_dma_convert_width(dws->n_bytes);
+	txconf.device_fc = false;
+
+	dmaengine_slave_config(dws->txchan, &txconf);
+
+	txdesc = dmaengine_prep_slave_sg(dws->txchan,
+				xfer->tx_sg.sgl,
+				xfer->tx_sg.nents,
+				DMA_MEM_TO_DEV,
+				DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!txdesc)
+		return NULL;
+
+	txdesc->callback = dw_spi_dma_tx_done;
+	txdesc->callback_param = dws;
+
+	return txdesc;
+}
+
+static inline bool dw_spi_dma_rx_busy(struct dw_spi *dws)
+{
+	return !!(dw_readl(dws, DW_SPI_SR) & SR_RF_NOT_EMPT);
+}
+
+static int dw_spi_dma_wait_rx_done(struct dw_spi *dws)
+{
+	int retry = WAIT_RETRIES;
+	struct spi_delay delay;
+	unsigned long ns, us;
+	u32 nents;
+
+	/*
+	 * It's unlikely that DMA engine is still doing the data fetching, but
+	 * if it's let's give it some reasonable time. The timeout calculation
+	 * is based on the synchronous APB/SSI reference clock rate, on a
+	 * number of data entries left in the Rx FIFO, times a number of clock
+	 * periods normally needed for a single APB read/write transaction
+	 * without PREADY signal utilized (which is true for the DW APB SSI
+	 * controller).
+	 */
+	nents = dw_readl(dws, DW_SPI_RXFLR);
+	ns = 4U * NSEC_PER_SEC / dws->max_freq * nents;
+	if (ns <= NSEC_PER_USEC) {
+		delay.unit = SPI_DELAY_UNIT_NSECS;
+		delay.value = ns;
+	} else {
+		us = DIV_ROUND_UP(ns, NSEC_PER_USEC);
+		delay.unit = SPI_DELAY_UNIT_USECS;
+		delay.value = clamp_val(us, 0, USHRT_MAX);
+	}
+
+	while (dw_spi_dma_rx_busy(dws) && retry--)
+		spi_delay_exec(&delay, NULL);
+
+	if (retry < 0) {
+		dev_err(&dws->master->dev, "Rx hanged up\n");
+		return -EIO;
+	}
+
+	return 0;
+}
+
+/*
+ * dws->dma_chan_busy is set before the dma transfer starts, callback for rx
+ * channel will clear a corresponding bit.
+ */
+static void dw_spi_dma_rx_done(void *arg)
+{
+	struct dw_spi *dws = arg;
+
+	clear_bit(RX_BUSY, &dws->dma_chan_busy);
+	if (test_bit(TX_BUSY, &dws->dma_chan_busy))
+		return;
+
+	dw_writel(dws, DW_SPI_DMACR, 0);
+	complete(&dws->dma_completion);
+}
+
+static struct dma_async_tx_descriptor *dw_spi_dma_prepare_rx(struct dw_spi *dws,
+		struct spi_transfer *xfer)
+{
+	struct dma_slave_config rxconf;
+	struct dma_async_tx_descriptor *rxdesc;
+
+	if (!xfer->rx_buf)
+		return NULL;
+
+	memset(&rxconf, 0, sizeof(rxconf));
+	rxconf.direction = DMA_DEV_TO_MEM;
+	rxconf.src_addr = dws->dma_addr;
+	rxconf.src_maxburst = dws->rxburst;
+	rxconf.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	rxconf.src_addr_width = dw_spi_dma_convert_width(dws->n_bytes);
+	rxconf.device_fc = false;
+
+	dmaengine_slave_config(dws->rxchan, &rxconf);
+
+	rxdesc = dmaengine_prep_slave_sg(dws->rxchan,
+				xfer->rx_sg.sgl,
+				xfer->rx_sg.nents,
+				DMA_DEV_TO_MEM,
+				DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!rxdesc)
+		return NULL;
+
+	rxdesc->callback = dw_spi_dma_rx_done;
+	rxdesc->callback_param = dws;
+
+	return rxdesc;
+}
+
+static int dw_spi_dma_setup(struct dw_spi *dws, struct spi_transfer *xfer)
+{
+	u16 imr = 0, dma_ctrl = 0;
+
+	dw_writel(dws, DW_SPI_DMARDLR, dws->rxburst - 1);
+	dw_writel(dws, DW_SPI_DMATDLR, dws->fifo_len - dws->txburst);
+
+	if (xfer->tx_buf)
+		dma_ctrl |= SPI_DMA_TDMAE;
+	if (xfer->rx_buf)
+		dma_ctrl |= SPI_DMA_RDMAE;
+	dw_writel(dws, DW_SPI_DMACR, dma_ctrl);
+
+	/* Set the interrupt mask */
+	if (xfer->tx_buf)
+		imr |= SPI_INT_TXOI;
+	if (xfer->rx_buf)
+		imr |= SPI_INT_RXUI | SPI_INT_RXOI;
+	spi_umask_intr(dws, imr);
+
+	reinit_completion(&dws->dma_completion);
+
+	dws->transfer_handler = dw_spi_dma_transfer_handler;
+
+	return 0;
+}
+
+static int dw_spi_dma_transfer(struct dw_spi *dws, struct spi_transfer *xfer)
+{
+	struct dma_async_tx_descriptor *txdesc, *rxdesc;
+	int ret;
+
+	/* Prepare the TX dma transfer */
+	txdesc = dw_spi_dma_prepare_tx(dws, xfer);
+
+	/* Prepare the RX dma transfer */
+	rxdesc = dw_spi_dma_prepare_rx(dws, xfer);
+
+	/* rx must be started before tx due to spi instinct */
+	if (rxdesc) {
+		set_bit(RX_BUSY, &dws->dma_chan_busy);
+		dmaengine_submit(rxdesc);
+		dma_async_issue_pending(dws->rxchan);
+	}
+
+	if (txdesc) {
+		set_bit(TX_BUSY, &dws->dma_chan_busy);
+		dmaengine_submit(txdesc);
+		dma_async_issue_pending(dws->txchan);
+	}
+
+	ret = dw_spi_dma_wait(dws, xfer);
+	if (ret)
+		return ret;
+
+	if (txdesc && dws->master->cur_msg->status == -EINPROGRESS) {
+		ret = dw_spi_dma_wait_tx_done(dws, xfer);
+		if (ret)
+			return ret;
+	}
+
+	if (rxdesc && dws->master->cur_msg->status == -EINPROGRESS)
+		ret = dw_spi_dma_wait_rx_done(dws);
+
+	return ret;
+}
+
+static void dw_spi_dma_stop(struct dw_spi *dws)
+{
+	if (test_bit(TX_BUSY, &dws->dma_chan_busy)) {
+		dmaengine_terminate_sync(dws->txchan);
+		clear_bit(TX_BUSY, &dws->dma_chan_busy);
+	}
+	if (test_bit(RX_BUSY, &dws->dma_chan_busy)) {
+		dmaengine_terminate_sync(dws->rxchan);
+		clear_bit(RX_BUSY, &dws->dma_chan_busy);
+	}
+
+	dw_writel(dws, DW_SPI_DMACR, 0);
+}
+
+static const struct dw_spi_dma_ops dw_spi_dma_mfld_ops = {
+	.dma_init	= dw_spi_dma_init_mfld,
+	.dma_exit	= dw_spi_dma_exit,
+	.dma_setup	= dw_spi_dma_setup,
+	.can_dma	= dw_spi_can_dma,
+	.dma_transfer	= dw_spi_dma_transfer,
+	.dma_stop	= dw_spi_dma_stop,
+};
+
+void dw_spi_dma_setup_mfld(struct dw_spi *dws)
+{
+	dws->dma_ops = &dw_spi_dma_mfld_ops;
+}
+EXPORT_SYMBOL_GPL(dw_spi_dma_setup_mfld);
+
+static const struct dw_spi_dma_ops dw_spi_dma_generic_ops = {
+	.dma_init	= dw_spi_dma_init_generic,
+	.dma_exit	= dw_spi_dma_exit,
+	.dma_setup	= dw_spi_dma_setup,
+	.can_dma	= dw_spi_can_dma,
+	.dma_transfer	= dw_spi_dma_transfer,
+	.dma_stop	= dw_spi_dma_stop,
+};
+
+void dw_spi_dma_setup_generic(struct dw_spi *dws)
+{
+	dws->dma_ops = &dw_spi_dma_generic_ops;
+}
+EXPORT_SYMBOL_GPL(dw_spi_dma_setup_generic);
diff --git a/drivers/spi/spi-dw-mid.c b/drivers/spi/spi-dw-mid.c
deleted file mode 100644
index 0d86c37e0aeb..000000000000
--- a/drivers/spi/spi-dw-mid.c
+++ /dev/null
@@ -1,322 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * Special handling for DW core on Intel MID platform
- *
- * Copyright (c) 2009, 2014 Intel Corporation.
- */
-
-#include <linux/dma-mapping.h>
-#include <linux/dmaengine.h>
-#include <linux/interrupt.h>
-#include <linux/slab.h>
-#include <linux/spi/spi.h>
-#include <linux/types.h>
-
-#include "spi-dw.h"
-
-#ifdef CONFIG_SPI_DW_MID_DMA
-#include <linux/pci.h>
-#include <linux/platform_data/dma-dw.h>
-
-#define RX_BUSY		0
-#define TX_BUSY		1
-
-static struct dw_dma_slave mid_dma_tx = { .dst_id = 1 };
-static struct dw_dma_slave mid_dma_rx = { .src_id = 0 };
-
-static bool mid_spi_dma_chan_filter(struct dma_chan *chan, void *param)
-{
-	struct dw_dma_slave *s = param;
-
-	if (s->dma_dev != chan->device->dev)
-		return false;
-
-	chan->private = s;
-	return true;
-}
-
-static int mid_spi_dma_init(struct dw_spi *dws)
-{
-	struct pci_dev *dma_dev;
-	struct dw_dma_slave *tx = dws->dma_tx;
-	struct dw_dma_slave *rx = dws->dma_rx;
-	dma_cap_mask_t mask;
-
-	/*
-	 * Get pci device for DMA controller, currently it could only
-	 * be the DMA controller of Medfield
-	 */
-	dma_dev = pci_get_device(PCI_VENDOR_ID_INTEL, 0x0827, NULL);
-	if (!dma_dev)
-		return -ENODEV;
-
-	dma_cap_zero(mask);
-	dma_cap_set(DMA_SLAVE, mask);
-
-	/* 1. Init rx channel */
-	rx->dma_dev = &dma_dev->dev;
-	dws->rxchan = dma_request_channel(mask, mid_spi_dma_chan_filter, rx);
-	if (!dws->rxchan)
-		goto err_exit;
-	dws->master->dma_rx = dws->rxchan;
-
-	/* 2. Init tx channel */
-	tx->dma_dev = &dma_dev->dev;
-	dws->txchan = dma_request_channel(mask, mid_spi_dma_chan_filter, tx);
-	if (!dws->txchan)
-		goto free_rxchan;
-	dws->master->dma_tx = dws->txchan;
-
-	dws->dma_inited = 1;
-	return 0;
-
-free_rxchan:
-	dma_release_channel(dws->rxchan);
-err_exit:
-	return -EBUSY;
-}
-
-static void mid_spi_dma_exit(struct dw_spi *dws)
-{
-	if (!dws->dma_inited)
-		return;
-
-	dmaengine_terminate_sync(dws->txchan);
-	dma_release_channel(dws->txchan);
-
-	dmaengine_terminate_sync(dws->rxchan);
-	dma_release_channel(dws->rxchan);
-}
-
-static irqreturn_t dma_transfer(struct dw_spi *dws)
-{
-	u16 irq_status = dw_readl(dws, DW_SPI_ISR);
-
-	if (!irq_status)
-		return IRQ_NONE;
-
-	dw_readl(dws, DW_SPI_ICR);
-	spi_reset_chip(dws);
-
-	dev_err(&dws->master->dev, "%s: FIFO overrun/underrun\n", __func__);
-	dws->master->cur_msg->status = -EIO;
-	spi_finalize_current_transfer(dws->master);
-	return IRQ_HANDLED;
-}
-
-static bool mid_spi_can_dma(struct spi_controller *master,
-		struct spi_device *spi, struct spi_transfer *xfer)
-{
-	struct dw_spi *dws = spi_controller_get_devdata(master);
-
-	if (!dws->dma_inited)
-		return false;
-
-	return xfer->len > dws->fifo_len;
-}
-
-static enum dma_slave_buswidth convert_dma_width(u32 dma_width) {
-	if (dma_width == 1)
-		return DMA_SLAVE_BUSWIDTH_1_BYTE;
-	else if (dma_width == 2)
-		return DMA_SLAVE_BUSWIDTH_2_BYTES;
-
-	return DMA_SLAVE_BUSWIDTH_UNDEFINED;
-}
-
-/*
- * dws->dma_chan_busy is set before the dma transfer starts, callback for tx
- * channel will clear a corresponding bit.
- */
-static void dw_spi_dma_tx_done(void *arg)
-{
-	struct dw_spi *dws = arg;
-
-	clear_bit(TX_BUSY, &dws->dma_chan_busy);
-	if (test_bit(RX_BUSY, &dws->dma_chan_busy))
-		return;
-	spi_finalize_current_transfer(dws->master);
-}
-
-static struct dma_async_tx_descriptor *dw_spi_dma_prepare_tx(struct dw_spi *dws,
-		struct spi_transfer *xfer)
-{
-	struct dma_slave_config txconf;
-	struct dma_async_tx_descriptor *txdesc;
-
-	if (!xfer->tx_buf)
-		return NULL;
-
-	txconf.direction = DMA_MEM_TO_DEV;
-	txconf.dst_addr = dws->dma_addr;
-	txconf.dst_maxburst = 16;
-	txconf.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
-	txconf.dst_addr_width = convert_dma_width(dws->dma_width);
-	txconf.device_fc = false;
-
-	dmaengine_slave_config(dws->txchan, &txconf);
-
-	txdesc = dmaengine_prep_slave_sg(dws->txchan,
-				xfer->tx_sg.sgl,
-				xfer->tx_sg.nents,
-				DMA_MEM_TO_DEV,
-				DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
-	if (!txdesc)
-		return NULL;
-
-	txdesc->callback = dw_spi_dma_tx_done;
-	txdesc->callback_param = dws;
-
-	return txdesc;
-}
-
-/*
- * dws->dma_chan_busy is set before the dma transfer starts, callback for rx
- * channel will clear a corresponding bit.
- */
-static void dw_spi_dma_rx_done(void *arg)
-{
-	struct dw_spi *dws = arg;
-
-	clear_bit(RX_BUSY, &dws->dma_chan_busy);
-	if (test_bit(TX_BUSY, &dws->dma_chan_busy))
-		return;
-	spi_finalize_current_transfer(dws->master);
-}
-
-static struct dma_async_tx_descriptor *dw_spi_dma_prepare_rx(struct dw_spi *dws,
-		struct spi_transfer *xfer)
-{
-	struct dma_slave_config rxconf;
-	struct dma_async_tx_descriptor *rxdesc;
-
-	if (!xfer->rx_buf)
-		return NULL;
-
-	rxconf.direction = DMA_DEV_TO_MEM;
-	rxconf.src_addr = dws->dma_addr;
-	rxconf.src_maxburst = 16;
-	rxconf.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
-	rxconf.src_addr_width = convert_dma_width(dws->dma_width);
-	rxconf.device_fc = false;
-
-	dmaengine_slave_config(dws->rxchan, &rxconf);
-
-	rxdesc = dmaengine_prep_slave_sg(dws->rxchan,
-				xfer->rx_sg.sgl,
-				xfer->rx_sg.nents,
-				DMA_DEV_TO_MEM,
-				DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
-	if (!rxdesc)
-		return NULL;
-
-	rxdesc->callback = dw_spi_dma_rx_done;
-	rxdesc->callback_param = dws;
-
-	return rxdesc;
-}
-
-static int mid_spi_dma_setup(struct dw_spi *dws, struct spi_transfer *xfer)
-{
-	u16 dma_ctrl = 0;
-
-	dw_writel(dws, DW_SPI_DMARDLR, 0xf);
-	dw_writel(dws, DW_SPI_DMATDLR, 0x10);
-
-	if (xfer->tx_buf)
-		dma_ctrl |= SPI_DMA_TDMAE;
-	if (xfer->rx_buf)
-		dma_ctrl |= SPI_DMA_RDMAE;
-	dw_writel(dws, DW_SPI_DMACR, dma_ctrl);
-
-	/* Set the interrupt mask */
-	spi_umask_intr(dws, SPI_INT_TXOI | SPI_INT_RXUI | SPI_INT_RXOI);
-
-	dws->transfer_handler = dma_transfer;
-
-	return 0;
-}
-
-static int mid_spi_dma_transfer(struct dw_spi *dws, struct spi_transfer *xfer)
-{
-	struct dma_async_tx_descriptor *txdesc, *rxdesc;
-
-	/* Prepare the TX dma transfer */
-	txdesc = dw_spi_dma_prepare_tx(dws, xfer);
-
-	/* Prepare the RX dma transfer */
-	rxdesc = dw_spi_dma_prepare_rx(dws, xfer);
-
-	/* rx must be started before tx due to spi instinct */
-	if (rxdesc) {
-		set_bit(RX_BUSY, &dws->dma_chan_busy);
-		dmaengine_submit(rxdesc);
-		dma_async_issue_pending(dws->rxchan);
-	}
-
-	if (txdesc) {
-		set_bit(TX_BUSY, &dws->dma_chan_busy);
-		dmaengine_submit(txdesc);
-		dma_async_issue_pending(dws->txchan);
-	}
-
-	return 0;
-}
-
-static void mid_spi_dma_stop(struct dw_spi *dws)
-{
-	if (test_bit(TX_BUSY, &dws->dma_chan_busy)) {
-		dmaengine_terminate_sync(dws->txchan);
-		clear_bit(TX_BUSY, &dws->dma_chan_busy);
-	}
-	if (test_bit(RX_BUSY, &dws->dma_chan_busy)) {
-		dmaengine_terminate_sync(dws->rxchan);
-		clear_bit(RX_BUSY, &dws->dma_chan_busy);
-	}
-}
-
-static const struct dw_spi_dma_ops mid_dma_ops = {
-	.dma_init	= mid_spi_dma_init,
-	.dma_exit	= mid_spi_dma_exit,
-	.dma_setup	= mid_spi_dma_setup,
-	.can_dma	= mid_spi_can_dma,
-	.dma_transfer	= mid_spi_dma_transfer,
-	.dma_stop	= mid_spi_dma_stop,
-};
-#endif
-
-/* Some specific info for SPI0 controller on Intel MID */
-
-/* HW info for MRST Clk Control Unit, 32b reg per controller */
-#define MRST_SPI_CLK_BASE	100000000	/* 100m */
-#define MRST_CLK_SPI_REG	0xff11d86c
-#define CLK_SPI_BDIV_OFFSET	0
-#define CLK_SPI_BDIV_MASK	0x00000007
-#define CLK_SPI_CDIV_OFFSET	9
-#define CLK_SPI_CDIV_MASK	0x00000e00
-#define CLK_SPI_DISABLE_OFFSET	8
-
-int dw_spi_mid_init(struct dw_spi *dws)
-{
-	void __iomem *clk_reg;
-	u32 clk_cdiv;
-
-	clk_reg = ioremap(MRST_CLK_SPI_REG, 16);
-	if (!clk_reg)
-		return -ENOMEM;
-
-	/* Get SPI controller operating freq info */
-	clk_cdiv = readl(clk_reg + dws->bus_num * sizeof(u32));
-	clk_cdiv &= CLK_SPI_CDIV_MASK;
-	clk_cdiv >>= CLK_SPI_CDIV_OFFSET;
-	dws->max_freq = MRST_SPI_CLK_BASE / (clk_cdiv + 1);
-
-	iounmap(clk_reg);
-
-#ifdef CONFIG_SPI_DW_MID_DMA
-	dws->dma_tx = &mid_dma_tx;
-	dws->dma_rx = &mid_dma_rx;
-	dws->dma_ops = &mid_dma_ops;
-#endif
-	return 0;
-}
diff --git a/drivers/spi/spi-dw-mmio.c b/drivers/spi/spi-dw-mmio.c
index 384a3ab6dc2d..403403deae66 100644
--- a/drivers/spi/spi-dw-mmio.c
+++ b/drivers/spi/spi-dw-mmio.c
@@ -7,7 +7,6 @@
 
 #include <linux/clk.h>
 #include <linux/err.h>
-#include <linux/interrupt.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/slab.h>
@@ -20,6 +19,7 @@
 #include <linux/acpi.h>
 #include <linux/property.h>
 #include <linux/regmap.h>
+#include <linux/reset.h>
 
 #include "spi-dw.h"
 
@@ -30,6 +30,7 @@ struct dw_spi_mmio {
 	struct clk     *clk;
 	struct clk     *pclk;
 	void           *priv;
+	struct reset_control *rstc;
 };
 
 #define MSCC_CPU_SYSTEM_CTRL_GENERAL_CTRL	0x24
@@ -44,6 +45,13 @@ struct dw_spi_mmio {
 #define MSCC_SPI_MST_SW_MODE_SW_PIN_CTRL_MODE	BIT(13)
 #define MSCC_SPI_MST_SW_MODE_SW_SPI_CS(x)	(x << 5)
 
+/*
+ * For Keem Bay, CTRLR0[31] is used to select controller mode.
+ * 0: SSI is slave
+ * 1: SSI is master
+ */
+#define KEEMBAY_CTRLR0_SSIC_IS_MST		BIT(31)
+
 struct dw_spi_mscc {
 	struct regmap       *syscon;
 	void __iomem        *spi_mst;
@@ -106,6 +114,9 @@ static int dw_spi_mscc_init(struct platform_device *pdev,
 	dwsmmio->dws.set_cs = dw_spi_mscc_set_cs;
 	dwsmmio->priv = dwsmscc;
 
+	/* Register hook to configure CTRLR0 */
+	dwsmmio->dws.update_cr0 = dw_spi_update_cr0;
+
 	return 0;
 }
 
@@ -128,6 +139,49 @@ static int dw_spi_alpine_init(struct platform_device *pdev,
 {
 	dwsmmio->dws.cs_override = 1;
 
+	/* Register hook to configure CTRLR0 */
+	dwsmmio->dws.update_cr0 = dw_spi_update_cr0;
+
+	return 0;
+}
+
+static int dw_spi_dw_apb_init(struct platform_device *pdev,
+			      struct dw_spi_mmio *dwsmmio)
+{
+	/* Register hook to configure CTRLR0 */
+	dwsmmio->dws.update_cr0 = dw_spi_update_cr0;
+
+	dw_spi_dma_setup_generic(&dwsmmio->dws);
+
+	return 0;
+}
+
+static int dw_spi_dwc_ssi_init(struct platform_device *pdev,
+			       struct dw_spi_mmio *dwsmmio)
+{
+	/* Register hook to configure CTRLR0 */
+	dwsmmio->dws.update_cr0 = dw_spi_update_cr0_v1_01a;
+
+	dw_spi_dma_setup_generic(&dwsmmio->dws);
+
+	return 0;
+}
+
+static u32 dw_spi_update_cr0_keembay(struct spi_controller *master,
+				     struct spi_device *spi,
+				     struct spi_transfer *transfer)
+{
+	u32 cr0 = dw_spi_update_cr0_v1_01a(master, spi, transfer);
+
+	return cr0 | KEEMBAY_CTRLR0_SSIC_IS_MST;
+}
+
+static int dw_spi_keembay_init(struct platform_device *pdev,
+			       struct dw_spi_mmio *dwsmmio)
+{
+	/* Register hook to configure CTRLR0 */
+	dwsmmio->dws.update_cr0 = dw_spi_update_cr0_keembay;
+
 	return 0;
 }
 
@@ -136,6 +190,7 @@ static int dw_spi_mmio_probe(struct platform_device *pdev)
 	int (*init_func)(struct platform_device *pdev,
 			 struct dw_spi_mmio *dwsmmio);
 	struct dw_spi_mmio *dwsmmio;
+	struct resource *mem;
 	struct dw_spi *dws;
 	int ret;
 	int num_cs;
@@ -148,11 +203,11 @@ static int dw_spi_mmio_probe(struct platform_device *pdev)
 	dws = &dwsmmio->dws;
 
 	/* Get basic io resource and map it */
-	dws->regs = devm_platform_ioremap_resource(pdev, 0);
-	if (IS_ERR(dws->regs)) {
-		dev_err(&pdev->dev, "SPI region map failed\n");
+	dws->regs = devm_platform_get_and_ioremap_resource(pdev, 0, &mem);
+	if (IS_ERR(dws->regs))
 		return PTR_ERR(dws->regs);
-	}
+
+	dws->paddr = mem->start;
 
 	dws->irq = platform_get_irq(pdev, 0);
 	if (dws->irq < 0)
@@ -175,6 +230,14 @@ static int dw_spi_mmio_probe(struct platform_device *pdev)
 	if (ret)
 		goto out_clk;
 
+	/* find an optional reset controller */
+	dwsmmio->rstc = devm_reset_control_get_optional_exclusive(&pdev->dev, "spi");
+	if (IS_ERR(dwsmmio->rstc)) {
+		ret = PTR_ERR(dwsmmio->rstc);
+		goto out_clk;
+	}
+	reset_control_deassert(dwsmmio->rstc);
+
 	dws->bus_num = pdev->id;
 
 	dws->max_freq = clk_get_rate(dwsmmio->clk);
@@ -208,6 +271,8 @@ out:
 	clk_disable_unprepare(dwsmmio->pclk);
 out_clk:
 	clk_disable_unprepare(dwsmmio->clk);
+	reset_control_assert(dwsmmio->rstc);
+
 	return ret;
 }
 
@@ -219,25 +284,30 @@ static int dw_spi_mmio_remove(struct platform_device *pdev)
 	pm_runtime_disable(&pdev->dev);
 	clk_disable_unprepare(dwsmmio->pclk);
 	clk_disable_unprepare(dwsmmio->clk);
+	reset_control_assert(dwsmmio->rstc);
 
 	return 0;
 }
 
 static const struct of_device_id dw_spi_mmio_of_match[] = {
-	{ .compatible = "snps,dw-apb-ssi", },
+	{ .compatible = "snps,dw-apb-ssi", .data = dw_spi_dw_apb_init},
 	{ .compatible = "mscc,ocelot-spi", .data = dw_spi_mscc_ocelot_init},
 	{ .compatible = "mscc,jaguar2-spi", .data = dw_spi_mscc_jaguar2_init},
 	{ .compatible = "amazon,alpine-dw-apb-ssi", .data = dw_spi_alpine_init},
-	{ .compatible = "renesas,rzn1-spi", },
+	{ .compatible = "renesas,rzn1-spi", .data = dw_spi_dw_apb_init},
+	{ .compatible = "snps,dwc-ssi-1.01a", .data = dw_spi_dwc_ssi_init},
+	{ .compatible = "intel,keembay-ssi", .data = dw_spi_keembay_init},
 	{ /* end of table */}
 };
 MODULE_DEVICE_TABLE(of, dw_spi_mmio_of_match);
 
+#ifdef CONFIG_ACPI
 static const struct acpi_device_id dw_spi_mmio_acpi_match[] = {
-	{"HISI0173", 0},
+	{"HISI0173", (kernel_ulong_t)dw_spi_dw_apb_init},
 	{},
 };
 MODULE_DEVICE_TABLE(acpi, dw_spi_mmio_acpi_match);
+#endif
 
 static struct platform_driver dw_spi_mmio_driver = {
 	.probe		= dw_spi_mmio_probe,
diff --git a/drivers/spi/spi-dw-pci.c b/drivers/spi/spi-dw-pci.c
index 12c131b5fb4e..2ea73809ca34 100644
--- a/drivers/spi/spi-dw-pci.c
+++ b/drivers/spi/spi-dw-pci.c
@@ -5,7 +5,6 @@
  * Copyright (c) 2009, 2014 Intel Corporation.
  */
 
-#include <linux/interrupt.h>
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
 #include <linux/slab.h>
@@ -16,6 +15,15 @@
 
 #define DRIVER_NAME "dw_spi_pci"
 
+/* HW info for MRST Clk Control Unit, 32b reg per controller */
+#define MRST_SPI_CLK_BASE	100000000	/* 100m */
+#define MRST_CLK_SPI_REG	0xff11d86c
+#define CLK_SPI_BDIV_OFFSET	0
+#define CLK_SPI_BDIV_MASK	0x00000007
+#define CLK_SPI_CDIV_OFFSET	9
+#define CLK_SPI_CDIV_MASK	0x00000e00
+#define CLK_SPI_DISABLE_OFFSET	8
+
 struct spi_pci_desc {
 	int	(*setup)(struct dw_spi *);
 	u16	num_cs;
@@ -23,19 +31,55 @@ struct spi_pci_desc {
 	u32	max_freq;
 };
 
+static int spi_mid_init(struct dw_spi *dws)
+{
+	void __iomem *clk_reg;
+	u32 clk_cdiv;
+
+	clk_reg = ioremap(MRST_CLK_SPI_REG, 16);
+	if (!clk_reg)
+		return -ENOMEM;
+
+	/* Get SPI controller operating freq info */
+	clk_cdiv = readl(clk_reg + dws->bus_num * sizeof(u32));
+	clk_cdiv &= CLK_SPI_CDIV_MASK;
+	clk_cdiv >>= CLK_SPI_CDIV_OFFSET;
+	dws->max_freq = MRST_SPI_CLK_BASE / (clk_cdiv + 1);
+
+	iounmap(clk_reg);
+
+	/* Register hook to configure CTRLR0 */
+	dws->update_cr0 = dw_spi_update_cr0;
+
+	dw_spi_dma_setup_mfld(dws);
+
+	return 0;
+}
+
+static int spi_generic_init(struct dw_spi *dws)
+{
+	/* Register hook to configure CTRLR0 */
+	dws->update_cr0 = dw_spi_update_cr0;
+
+	dw_spi_dma_setup_generic(dws);
+
+	return 0;
+}
+
 static struct spi_pci_desc spi_pci_mid_desc_1 = {
-	.setup = dw_spi_mid_init,
+	.setup = spi_mid_init,
 	.num_cs = 5,
 	.bus_num = 0,
 };
 
 static struct spi_pci_desc spi_pci_mid_desc_2 = {
-	.setup = dw_spi_mid_init,
+	.setup = spi_mid_init,
 	.num_cs = 2,
 	.bus_num = 1,
 };
 
 static struct spi_pci_desc spi_pci_ehl_desc = {
+	.setup = spi_generic_init,
 	.num_cs = 2,
 	.bus_num = -1,
 	.max_freq = 100000000,
diff --git a/drivers/spi/spi-dw.h b/drivers/spi/spi-dw.h
index 1bf5713e047d..151ba316619e 100644
--- a/drivers/spi/spi-dw.h
+++ b/drivers/spi/spi-dw.h
@@ -2,18 +2,21 @@
 #ifndef DW_SPI_HEADER_H
 #define DW_SPI_HEADER_H
 
+#include <linux/completion.h>
+#include <linux/debugfs.h>
+#include <linux/irqreturn.h>
 #include <linux/io.h>
 #include <linux/scatterlist.h>
 
 /* Register offsets */
-#define DW_SPI_CTRL0			0x00
-#define DW_SPI_CTRL1			0x04
+#define DW_SPI_CTRLR0			0x00
+#define DW_SPI_CTRLR1			0x04
 #define DW_SPI_SSIENR			0x08
 #define DW_SPI_MWCR			0x0c
 #define DW_SPI_SER			0x10
 #define DW_SPI_BAUDR			0x14
-#define DW_SPI_TXFLTR			0x18
-#define DW_SPI_RXFLTR			0x1c
+#define DW_SPI_TXFTLR			0x18
+#define DW_SPI_RXFTLR			0x1c
 #define DW_SPI_TXFLR			0x20
 #define DW_SPI_RXFLR			0x24
 #define DW_SPI_SR			0x28
@@ -57,6 +60,15 @@
 #define SPI_SRL_OFFSET			11
 #define SPI_CFS_OFFSET			12
 
+/* Bit fields in CTRLR0 based on DWC_ssi_databook.pdf v1.01a */
+#define DWC_SSI_CTRLR0_SRL_OFFSET	13
+#define DWC_SSI_CTRLR0_TMOD_OFFSET	10
+#define DWC_SSI_CTRLR0_TMOD_MASK	GENMASK(11, 10)
+#define DWC_SSI_CTRLR0_SCPOL_OFFSET	9
+#define DWC_SSI_CTRLR0_SCPH_OFFSET	8
+#define DWC_SSI_CTRLR0_FRF_OFFSET	6
+#define DWC_SSI_CTRLR0_DFS_OFFSET	0
+
 /* Bit fields in SR, 7 bits */
 #define SR_MASK				0x7f		/* cover 7 bits */
 #define SR_BUSY				(1 << 0)
@@ -90,7 +102,7 @@ enum dw_ssi_type {
 
 struct dw_spi;
 struct dw_spi_dma_ops {
-	int (*dma_init)(struct dw_spi *dws);
+	int (*dma_init)(struct device *dev, struct dw_spi *dws);
 	void (*dma_exit)(struct dw_spi *dws);
 	int (*dma_setup)(struct dw_spi *dws, struct spi_transfer *xfer);
 	bool (*can_dma)(struct spi_controller *master, struct spi_device *spi,
@@ -114,6 +126,8 @@ struct dw_spi {
 	u16			bus_num;
 	u16			num_cs;		/* supported slave numbers */
 	void (*set_cs)(struct spi_device *spi, bool enable);
+	u32 (*update_cr0)(struct spi_controller *master, struct spi_device *spi,
+			  struct spi_transfer *transfer);
 
 	/* Current message transfer state info */
 	size_t			len;
@@ -124,24 +138,22 @@ struct dw_spi {
 	void			*rx_end;
 	int			dma_mapped;
 	u8			n_bytes;	/* current is a 1/2 bytes op */
-	u32			dma_width;
 	irqreturn_t		(*transfer_handler)(struct dw_spi *dws);
 	u32			current_freq;	/* frequency in hz */
 
 	/* DMA info */
-	int			dma_inited;
 	struct dma_chan		*txchan;
+	u32			txburst;
 	struct dma_chan		*rxchan;
+	u32			rxburst;
 	unsigned long		dma_chan_busy;
 	dma_addr_t		dma_addr; /* phy address of the Data register */
 	const struct dw_spi_dma_ops *dma_ops;
-	void			*dma_tx;
-	void			*dma_rx;
+	struct completion	dma_completion;
 
-	/* Bus interface info */
-	void			*priv;
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *debugfs;
+	struct debugfs_regset32 regset;
 #endif
 };
 
@@ -235,24 +247,28 @@ static inline void spi_shutdown_chip(struct dw_spi *dws)
 	spi_set_clk(dws, 0);
 }
 
-/*
- * Each SPI slave device to work with dw_api controller should
- * has such a structure claiming its working mode (poll or PIO/DMA),
- * which can be save in the "controller_data" member of the
- * struct spi_device.
- */
-struct dw_spi_chip {
-	u8 poll_mode;	/* 1 for controller polling mode */
-	u8 type;	/* SPI/SSP/MicroWire */
-	void (*cs_control)(u32 command);
-};
-
 extern void dw_spi_set_cs(struct spi_device *spi, bool enable);
 extern int dw_spi_add_host(struct device *dev, struct dw_spi *dws);
 extern void dw_spi_remove_host(struct dw_spi *dws);
 extern int dw_spi_suspend_host(struct dw_spi *dws);
 extern int dw_spi_resume_host(struct dw_spi *dws);
+extern u32 dw_spi_update_cr0(struct spi_controller *master,
+			     struct spi_device *spi,
+			     struct spi_transfer *transfer);
+extern u32 dw_spi_update_cr0_v1_01a(struct spi_controller *master,
+				    struct spi_device *spi,
+				    struct spi_transfer *transfer);
+
+#ifdef CONFIG_SPI_DW_DMA
+
+extern void dw_spi_dma_setup_mfld(struct dw_spi *dws);
+extern void dw_spi_dma_setup_generic(struct dw_spi *dws);
+
+#else
+
+static inline void dw_spi_dma_setup_mfld(struct dw_spi *dws) {}
+static inline void dw_spi_dma_setup_generic(struct dw_spi *dws) {}
+
+#endif /* !CONFIG_SPI_DW_DMA */
 
-/* platform related setup */
-extern int dw_spi_mid_init(struct dw_spi *dws); /* Intel MID platforms */
 #endif /* DW_SPI_HEADER_H */
diff --git a/drivers/spi/spi-ep93xx.c b/drivers/spi/spi-ep93xx.c
index 4e1ccd4e52b6..8c854b187b1d 100644
--- a/drivers/spi/spi-ep93xx.c
+++ b/drivers/spi/spi-ep93xx.c
@@ -31,7 +31,8 @@
 #include <linux/platform_data/spi-ep93xx.h>
 
 #define SSPCR0			0x0000
-#define SSPCR0_MODE_SHIFT	6
+#define SSPCR0_SPO		BIT(6)
+#define SSPCR0_SPH		BIT(7)
 #define SSPCR0_SCR_SHIFT	8
 
 #define SSPCR1			0x0004
@@ -159,7 +160,10 @@ static int ep93xx_spi_chip_setup(struct spi_master *master,
 		return err;
 
 	cr0 = div_scr << SSPCR0_SCR_SHIFT;
-	cr0 |= (spi->mode & (SPI_CPHA | SPI_CPOL)) << SSPCR0_MODE_SHIFT;
+	if (spi->mode & SPI_CPOL)
+		cr0 |= SSPCR0_SPO;
+	if (spi->mode & SPI_CPHA)
+		cr0 |= SSPCR0_SPH;
 	cr0 |= dss;
 
 	dev_dbg(&master->dev, "setup: mode %d, cpsr %d, scr %d, dss %d\n",
diff --git a/drivers/spi/spi-fsl-dspi.c b/drivers/spi/spi-fsl-dspi.c
index 50e41f66a2d7..a35faced0456 100644
--- a/drivers/spi/spi-fsl-dspi.c
+++ b/drivers/spi/spi-fsl-dspi.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0+
 //
 // Copyright 2013 Freescale Semiconductor, Inc.
+// Copyright 2020 NXP
 //
 // Freescale DSPI driver
 // This file contains a driver for the Freescale DSPI
@@ -26,6 +27,9 @@
 #define SPI_MCR_CLR_TXF			BIT(11)
 #define SPI_MCR_CLR_RXF			BIT(10)
 #define SPI_MCR_XSPI			BIT(3)
+#define SPI_MCR_DIS_TXF			BIT(13)
+#define SPI_MCR_DIS_RXF			BIT(12)
+#define SPI_MCR_HALT			BIT(0)
 
 #define SPI_TCR				0x08
 #define SPI_TCR_GET_TCNT(x)		(((x) & GENMASK(31, 16)) >> 16)
@@ -246,13 +250,33 @@ struct fsl_dspi {
 
 static void dspi_native_host_to_dev(struct fsl_dspi *dspi, u32 *txdata)
 {
-	memcpy(txdata, dspi->tx, dspi->oper_word_size);
+	switch (dspi->oper_word_size) {
+	case 1:
+		*txdata = *(u8 *)dspi->tx;
+		break;
+	case 2:
+		*txdata = *(u16 *)dspi->tx;
+		break;
+	case 4:
+		*txdata = *(u32 *)dspi->tx;
+		break;
+	}
 	dspi->tx += dspi->oper_word_size;
 }
 
 static void dspi_native_dev_to_host(struct fsl_dspi *dspi, u32 rxdata)
 {
-	memcpy(dspi->rx, &rxdata, dspi->oper_word_size);
+	switch (dspi->oper_word_size) {
+	case 1:
+		*(u8 *)dspi->rx = rxdata;
+		break;
+	case 2:
+		*(u16 *)dspi->rx = rxdata;
+		break;
+	case 4:
+		*(u32 *)dspi->rx = rxdata;
+		break;
+	}
 	dspi->rx += dspi->oper_word_size;
 }
 
@@ -1417,6 +1441,24 @@ static int dspi_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static void dspi_shutdown(struct platform_device *pdev)
+{
+	struct spi_controller *ctlr = platform_get_drvdata(pdev);
+	struct fsl_dspi *dspi = spi_controller_get_devdata(ctlr);
+
+	/* Disable RX and TX */
+	regmap_update_bits(dspi->regmap, SPI_MCR,
+			   SPI_MCR_DIS_TXF | SPI_MCR_DIS_RXF,
+			   SPI_MCR_DIS_TXF | SPI_MCR_DIS_RXF);
+
+	/* Stop Running */
+	regmap_update_bits(dspi->regmap, SPI_MCR, SPI_MCR_HALT, SPI_MCR_HALT);
+
+	dspi_release_dma(dspi);
+	clk_disable_unprepare(dspi->clk);
+	spi_unregister_controller(dspi->ctlr);
+}
+
 static struct platform_driver fsl_dspi_driver = {
 	.driver.name		= DRIVER_NAME,
 	.driver.of_match_table	= fsl_dspi_dt_ids,
@@ -1424,6 +1466,7 @@ static struct platform_driver fsl_dspi_driver = {
 	.driver.pm		= &dspi_pm,
 	.probe			= dspi_probe,
 	.remove			= dspi_remove,
+	.shutdown		= dspi_shutdown,
 };
 module_platform_driver(fsl_dspi_driver);
 
diff --git a/drivers/spi/spi-fsl-lpspi.c b/drivers/spi/spi-fsl-lpspi.c
index 8b41b70f6f5c..1552b28b9515 100644
--- a/drivers/spi/spi-fsl-lpspi.c
+++ b/drivers/spi/spi-fsl-lpspi.c
@@ -186,14 +186,13 @@ static bool fsl_lpspi_can_dma(struct spi_controller *controller,
 
 	bytes_per_word = fsl_lpspi_bytes_per_word(transfer->bits_per_word);
 
-	switch (bytes_per_word)
-	{
-		case 1:
-		case 2:
-		case 4:
-			break;
-		default:
-			return false;
+	switch (bytes_per_word) {
+	case 1:
+	case 2:
+	case 4:
+		break;
+	default:
+		return false;
 	}
 
 	return true;
@@ -941,7 +940,7 @@ static int fsl_lpspi_probe(struct platform_device *pdev)
 	ret = pm_runtime_get_sync(fsl_lpspi->dev);
 	if (ret < 0) {
 		dev_err(fsl_lpspi->dev, "failed to enable clock\n");
-		goto out_controller_put;
+		goto out_pm_get;
 	}
 
 	temp = readl(fsl_lpspi->base + IMX7ULP_PARAM);
@@ -950,13 +949,15 @@ static int fsl_lpspi_probe(struct platform_device *pdev)
 
 	ret = fsl_lpspi_dma_init(&pdev->dev, fsl_lpspi, controller);
 	if (ret == -EPROBE_DEFER)
-		goto out_controller_put;
+		goto out_pm_get;
 
 	if (ret < 0)
 		dev_err(&pdev->dev, "dma setup error %d, use pio\n", ret);
 
 	return 0;
 
+out_pm_get:
+	pm_runtime_put_noidle(fsl_lpspi->dev);
 out_controller_put:
 	spi_controller_put(controller);
 
diff --git a/drivers/spi/spi-fsl-qspi.c b/drivers/spi/spi-fsl-qspi.c
index 02e5cba0a5bb..6766262d7e75 100644
--- a/drivers/spi/spi-fsl-qspi.c
+++ b/drivers/spi/spi-fsl-qspi.c
@@ -876,14 +876,15 @@ static int fsl_qspi_probe(struct platform_device *pdev)
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
 					"QuadSPI-memory");
-	q->ahb_addr = devm_ioremap_resource(dev, res);
-	if (IS_ERR(q->ahb_addr)) {
-		ret = PTR_ERR(q->ahb_addr);
+	q->memmap_phy = res->start;
+	/* Since there are 4 cs, map size required is 4 times ahb_buf_size */
+	q->ahb_addr = devm_ioremap(dev, q->memmap_phy,
+				   (q->devtype_data->ahb_buf_size * 4));
+	if (!q->ahb_addr) {
+		ret = -ENOMEM;
 		goto err_put_ctrl;
 	}
 
-	q->memmap_phy = res->start;
-
 	/* find the clocks */
 	q->clk_en = devm_clk_get(dev, "qspi_en");
 	if (IS_ERR(q->clk_en)) {
diff --git a/drivers/spi/spi-fsl-spi.c b/drivers/spi/spi-fsl-spi.c
index 3b81772fea0d..67f022b8c81d 100644
--- a/drivers/spi/spi-fsl-spi.c
+++ b/drivers/spi/spi-fsl-spi.c
@@ -588,7 +588,7 @@ static void fsl_spi_grlib_probe(struct device *dev)
 	pdata->cs_control = fsl_spi_grlib_cs_control;
 }
 
-static struct spi_master * fsl_spi_probe(struct device *dev,
+static struct spi_master *fsl_spi_probe(struct device *dev,
 		struct resource *mem, unsigned int irq)
 {
 	struct fsl_spi_platform_data *pdata = dev_get_platdata(dev);
diff --git a/drivers/spi/spi-hisi-sfc-v3xx.c b/drivers/spi/spi-hisi-sfc-v3xx.c
index e3b57252d075..64a18d08a4d9 100644
--- a/drivers/spi/spi-hisi-sfc-v3xx.c
+++ b/drivers/spi/spi-hisi-sfc-v3xx.c
@@ -17,6 +17,11 @@
 
 #define HISI_SFC_V3XX_VERSION (0x1f8)
 
+#define HISI_SFC_V3XX_INT_STAT (0x120)
+#define HISI_SFC_V3XX_INT_STAT_PP_ERR BIT(2)
+#define HISI_SFC_V3XX_INT_STAT_ADDR_IACCES BIT(5)
+#define HISI_SFC_V3XX_INT_CLR (0x12c)
+#define HISI_SFC_V3XX_INT_CLR_CLEAR (0xff)
 #define HISI_SFC_V3XX_CMD_CFG (0x300)
 #define HISI_SFC_V3XX_CMD_CFG_DUAL_IN_DUAL_OUT (1 << 17)
 #define HISI_SFC_V3XX_CMD_CFG_DUAL_IO (2 << 17)
@@ -163,7 +168,7 @@ static int hisi_sfc_v3xx_generic_exec_op(struct hisi_sfc_v3xx_host *host,
 					 u8 chip_select)
 {
 	int ret, len = op->data.nbytes;
-	u32 config = 0;
+	u32 int_stat, config = 0;
 
 	if (op->addr.nbytes)
 		config |= HISI_SFC_V3XX_CMD_CFG_ADDR_EN_MSK;
@@ -228,6 +233,25 @@ static int hisi_sfc_v3xx_generic_exec_op(struct hisi_sfc_v3xx_host *host,
 	if (ret)
 		return ret;
 
+	/*
+	 * The interrupt status register indicates whether an error occurs
+	 * after per operation. Check it, and clear the interrupts for
+	 * next time judgement.
+	 */
+	int_stat = readl(host->regbase + HISI_SFC_V3XX_INT_STAT);
+	writel(HISI_SFC_V3XX_INT_CLR_CLEAR,
+	       host->regbase + HISI_SFC_V3XX_INT_CLR);
+
+	if (int_stat & HISI_SFC_V3XX_INT_STAT_ADDR_IACCES) {
+		dev_err(host->dev, "fail to access protected address\n");
+		return -EIO;
+	}
+
+	if (int_stat & HISI_SFC_V3XX_INT_STAT_PP_ERR) {
+		dev_err(host->dev, "page program operation failed\n");
+		return -EIO;
+	}
+
 	if (op->data.dir == SPI_MEM_DATA_IN)
 		hisi_sfc_v3xx_read_databuf(host, op->data.buf.in, len);
 
diff --git a/drivers/spi/spi-imx.c b/drivers/spi/spi-imx.c
index f4f28a400a96..b7a85e3fe1c1 100644
--- a/drivers/spi/spi-imx.c
+++ b/drivers/spi/spi-imx.c
@@ -71,6 +71,7 @@ struct spi_imx_devtype_data {
 	void (*reset)(struct spi_imx_data *);
 	void (*setup_wml)(struct spi_imx_data *);
 	void (*disable)(struct spi_imx_data *);
+	void (*disable_dma)(struct spi_imx_data *);
 	bool has_dmamode;
 	bool has_slavemode;
 	unsigned int fifo_size;
@@ -485,6 +486,11 @@ static void mx51_ecspi_trigger(struct spi_imx_data *spi_imx)
 	writel(reg, spi_imx->base + MX51_ECSPI_CTRL);
 }
 
+static void mx51_disable_dma(struct spi_imx_data *spi_imx)
+{
+	writel(0, spi_imx->base + MX51_ECSPI_DMA);
+}
+
 static void mx51_ecspi_disable(struct spi_imx_data *spi_imx)
 {
 	u32 ctrl;
@@ -987,6 +993,7 @@ static struct spi_imx_devtype_data imx51_ecspi_devtype_data = {
 	.rx_available = mx51_ecspi_rx_available,
 	.reset = mx51_ecspi_reset,
 	.setup_wml = mx51_setup_wml,
+	.disable_dma = mx51_disable_dma,
 	.fifo_size = 64,
 	.has_dmamode = true,
 	.dynamic_burst = true,
@@ -1001,6 +1008,7 @@ static struct spi_imx_devtype_data imx53_ecspi_devtype_data = {
 	.prepare_transfer = mx51_ecspi_prepare_transfer,
 	.trigger = mx51_ecspi_trigger,
 	.rx_available = mx51_ecspi_rx_available,
+	.disable_dma = mx51_disable_dma,
 	.reset = mx51_ecspi_reset,
 	.fifo_size = 64,
 	.has_dmamode = true,
@@ -1385,6 +1393,7 @@ static int spi_imx_dma_transfer(struct spi_imx_data *spi_imx,
 				DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 	if (!desc_tx) {
 		dmaengine_terminate_all(master->dma_tx);
+		dmaengine_terminate_all(master->dma_rx);
 		return -EINVAL;
 	}
 
@@ -1498,6 +1507,7 @@ static int spi_imx_transfer(struct spi_device *spi,
 				struct spi_transfer *transfer)
 {
 	struct spi_imx_data *spi_imx = spi_master_get_devdata(spi->master);
+	int ret;
 
 	/* flush rxfifo before transfer */
 	while (spi_imx->devtype_data->rx_available(spi_imx))
@@ -1506,10 +1516,23 @@ static int spi_imx_transfer(struct spi_device *spi,
 	if (spi_imx->slave_mode)
 		return spi_imx_pio_transfer_slave(spi, transfer);
 
-	if (spi_imx->usedma)
-		return spi_imx_dma_transfer(spi_imx, transfer);
-	else
-		return spi_imx_pio_transfer(spi, transfer);
+	/*
+	 * fallback PIO mode if dma setup error happen, for example sdma
+	 * firmware may not be updated as ERR009165 required.
+	 */
+	if (spi_imx->usedma) {
+		ret = spi_imx_dma_transfer(spi_imx, transfer);
+		if (ret != -EINVAL)
+			return ret;
+
+		spi_imx->devtype_data->disable_dma(spi_imx);
+
+		spi_imx->usedma = false;
+		spi_imx->dynamic_burst = spi_imx->devtype_data->dynamic_burst;
+		dev_dbg(&spi->dev, "Fallback to PIO mode\n");
+	}
+
+	return spi_imx_pio_transfer(spi, transfer);
 }
 
 static int spi_imx_setup(struct spi_device *spi)
diff --git a/drivers/spi/spi-mem.c b/drivers/spi/spi-mem.c
index adaa0c49f966..9a86cc27fcc0 100644
--- a/drivers/spi/spi-mem.c
+++ b/drivers/spi/spi-mem.c
@@ -108,15 +108,17 @@ static int spi_check_buswidth_req(struct spi_mem *mem, u8 buswidth, bool tx)
 		return 0;
 
 	case 2:
-		if ((tx && (mode & (SPI_TX_DUAL | SPI_TX_QUAD))) ||
-		    (!tx && (mode & (SPI_RX_DUAL | SPI_RX_QUAD))))
+		if ((tx &&
+		     (mode & (SPI_TX_DUAL | SPI_TX_QUAD | SPI_TX_OCTAL))) ||
+		    (!tx &&
+		     (mode & (SPI_RX_DUAL | SPI_RX_QUAD | SPI_RX_OCTAL))))
 			return 0;
 
 		break;
 
 	case 4:
-		if ((tx && (mode & SPI_TX_QUAD)) ||
-		    (!tx && (mode & SPI_RX_QUAD)))
+		if ((tx && (mode & (SPI_TX_QUAD | SPI_TX_OCTAL))) ||
+		    (!tx && (mode & (SPI_RX_QUAD | SPI_RX_OCTAL))))
 			return 0;
 
 		break;
diff --git a/drivers/spi/spi-mtk-nor.c b/drivers/spi/spi-mtk-nor.c
index c15a9910549f..7bc302b50396 100644
--- a/drivers/spi/spi-mtk-nor.c
+++ b/drivers/spi/spi-mtk-nor.c
@@ -391,7 +391,7 @@ static int mtk_nor_pp_unbuffered(struct mtk_nor *sp,
 	return mtk_nor_cmd_exec(sp, MTK_NOR_CMD_WRITE, 6 * BITS_PER_BYTE);
 }
 
-int mtk_nor_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
+static int mtk_nor_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
 {
 	struct mtk_nor *sp = spi_controller_get_devdata(mem->spi->master);
 	int ret;
diff --git a/drivers/spi/spi-mux.c b/drivers/spi/spi-mux.c
index 4f94c9127fc1..cc9ef371db14 100644
--- a/drivers/spi/spi-mux.c
+++ b/drivers/spi/spi-mux.c
@@ -51,6 +51,10 @@ static int spi_mux_select(struct spi_device *spi)
 	struct spi_mux_priv *priv = spi_controller_get_devdata(spi->controller);
 	int ret;
 
+	ret = mux_control_select(priv->mux, spi->chip_select);
+	if (ret)
+		return ret;
+
 	if (priv->current_cs == spi->chip_select)
 		return 0;
 
@@ -62,10 +66,6 @@ static int spi_mux_select(struct spi_device *spi)
 	priv->spi->mode = spi->mode;
 	priv->spi->bits_per_word = spi->bits_per_word;
 
-	ret = mux_control_select(priv->mux, spi->chip_select);
-	if (ret)
-		return ret;
-
 	priv->current_cs = spi->chip_select;
 
 	return 0;
diff --git a/drivers/spi/spi-orion.c b/drivers/spi/spi-orion.c
index 1f59beb7d27e..43f73db22f21 100644
--- a/drivers/spi/spi-orion.c
+++ b/drivers/spi/spi-orion.c
@@ -17,10 +17,8 @@
 #include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_device.h>
-#include <linux/of_gpio.h>
 #include <linux/clk.h>
 #include <linux/sizes.h>
-#include <linux/gpio.h>
 #include <asm/unaligned.h>
 
 #define DRIVER_NAME			"orion_spi"
@@ -98,7 +96,6 @@ struct orion_spi {
 	struct clk              *clk;
 	struct clk              *axi_clk;
 	const struct orion_spi_dev *devdata;
-	int			unused_hw_gpio;
 
 	struct orion_child_options	child[ORION_NUM_CHIPSELECTS];
 };
@@ -325,20 +322,27 @@ orion_spi_setup_transfer(struct spi_device *spi, struct spi_transfer *t)
 static void orion_spi_set_cs(struct spi_device *spi, bool enable)
 {
 	struct orion_spi *orion_spi;
-	int cs;
 
 	orion_spi = spi_master_get_devdata(spi->master);
 
-	if (gpio_is_valid(spi->cs_gpio))
-		cs = orion_spi->unused_hw_gpio;
-	else
-		cs = spi->chip_select;
-
+	/*
+	 * If this line is using a GPIO to control chip select, this internal
+	 * .set_cs() function will still be called, so we clear any previous
+	 * chip select. The CS we activate will not have any elecrical effect,
+	 * as it is handled by a GPIO, but that doesn't matter. What we need
+	 * is to deassert the old chip select and assert some other chip select.
+	 */
 	orion_spi_clrbits(orion_spi, ORION_SPI_IF_CTRL_REG, ORION_SPI_CS_MASK);
 	orion_spi_setbits(orion_spi, ORION_SPI_IF_CTRL_REG,
-				ORION_SPI_CS(cs));
+			  ORION_SPI_CS(spi->chip_select));
 
-	/* Chip select logic is inverted from spi_set_cs */
+	/*
+	 * Chip select logic is inverted from spi_set_cs(). For lines using a
+	 * GPIO to do chip select SPI_CS_HIGH is enforced and inversion happens
+	 * in the GPIO library, but we don't care about that, because in those
+	 * cases we are dealing with an unused native CS anyways so the polarity
+	 * doesn't matter.
+	 */
 	if (!enable)
 		orion_spi_setbits(orion_spi, ORION_SPI_IF_CTRL_REG, 0x1);
 	else
@@ -503,9 +507,6 @@ static int orion_spi_transfer_one(struct spi_master *master,
 
 static int orion_spi_setup(struct spi_device *spi)
 {
-	if (gpio_is_valid(spi->cs_gpio)) {
-		gpio_direction_output(spi->cs_gpio, !(spi->mode & SPI_CS_HIGH));
-	}
 	return orion_spi_setup_transfer(spi, NULL);
 }
 
@@ -622,13 +623,13 @@ static int orion_spi_probe(struct platform_device *pdev)
 	master->setup = orion_spi_setup;
 	master->bits_per_word_mask = SPI_BPW_MASK(8) | SPI_BPW_MASK(16);
 	master->auto_runtime_pm = true;
+	master->use_gpio_descriptors = true;
 	master->flags = SPI_MASTER_GPIO_SS;
 
 	platform_set_drvdata(pdev, master);
 
 	spi = spi_master_get_devdata(master);
 	spi->master = master;
-	spi->unused_hw_gpio = -1;
 
 	of_id = of_match_device(orion_spi_of_match_table, &pdev->dev);
 	devdata = (of_id) ? of_id->data : &orion_spi_dev_data;
@@ -683,7 +684,6 @@ static int orion_spi_probe(struct platform_device *pdev)
 	for_each_available_child_of_node(pdev->dev.of_node, np) {
 		struct orion_direct_acc *dir_acc;
 		u32 cs;
-		int cs_gpio;
 
 		/* Get chip-select number from the "reg" property */
 		status = of_property_read_u32(np, "reg", &cs);
@@ -695,44 +695,6 @@ static int orion_spi_probe(struct platform_device *pdev)
 		}
 
 		/*
-		 * Initialize the CS GPIO:
-		 * - properly request the actual GPIO signal
-		 * - de-assert the logical signal so that all GPIO CS lines
-		 *   are inactive when probing for slaves
-		 * - find an unused physical CS which will be driven for any
-		 *   slave which uses a CS GPIO
-		 */
-		cs_gpio = of_get_named_gpio(pdev->dev.of_node, "cs-gpios", cs);
-		if (cs_gpio > 0) {
-			char *gpio_name;
-			int cs_flags;
-
-			if (spi->unused_hw_gpio == -1) {
-				dev_info(&pdev->dev,
-					"Selected unused HW CS#%d for any GPIO CSes\n",
-					cs);
-				spi->unused_hw_gpio = cs;
-			}
-
-			gpio_name = devm_kasprintf(&pdev->dev, GFP_KERNEL,
-					"%s-CS%d", dev_name(&pdev->dev), cs);
-			if (!gpio_name) {
-				status = -ENOMEM;
-				goto out_rel_axi_clk;
-			}
-
-			cs_flags = of_property_read_bool(np, "spi-cs-high") ?
-				GPIOF_OUT_INIT_LOW : GPIOF_OUT_INIT_HIGH;
-			status = devm_gpio_request_one(&pdev->dev, cs_gpio,
-					cs_flags, gpio_name);
-			if (status) {
-				dev_err(&pdev->dev,
-					"Can't request GPIO for CS %d\n", cs);
-				goto out_rel_axi_clk;
-			}
-		}
-
-		/*
 		 * Check if an address is configured for this SPI device. If
 		 * not, the MBus mapping via the 'ranges' property in the 'soc'
 		 * node is not configured and this device should not use the
diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c
index 73d2a65d0b6e..6721910e5f2a 100644
--- a/drivers/spi/spi-pxa2xx.c
+++ b/drivers/spi/spi-pxa2xx.c
@@ -150,6 +150,7 @@ static const struct lpss_config lpss_platforms[] = {
 		.tx_threshold_hi = 48,
 		.cs_sel_shift = 8,
 		.cs_sel_mask = 3 << 8,
+		.cs_clk_stays_gated = true,
 	},
 	{	/* LPSS_CNL_SSP */
 		.offset = 0x200,
@@ -1884,7 +1885,7 @@ static int pxa2xx_spi_probe(struct platform_device *pdev)
 
 	/* Register with the SPI framework */
 	platform_set_drvdata(pdev, drv_data);
-	status = devm_spi_register_controller(&pdev->dev, controller);
+	status = spi_register_controller(controller);
 	if (status != 0) {
 		dev_err(&pdev->dev, "problem registering spi controller\n");
 		goto out_error_pm_runtime_enabled;
@@ -1893,7 +1894,6 @@ static int pxa2xx_spi_probe(struct platform_device *pdev)
 	return status;
 
 out_error_pm_runtime_enabled:
-	pm_runtime_put_noidle(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 
 out_error_clock_enabled:
@@ -1916,6 +1916,8 @@ static int pxa2xx_spi_remove(struct platform_device *pdev)
 
 	pm_runtime_get_sync(&pdev->dev);
 
+	spi_unregister_controller(drv_data->controller);
+
 	/* Disable the SSP at the peripheral and SOC level */
 	pxa2xx_spi_write(drv_data, SSCR0, 0);
 	clk_disable_unprepare(ssp->clk);
diff --git a/drivers/spi/spi-rb4xx.c b/drivers/spi/spi-rb4xx.c
index 4c9620e0d18c..8aa51beb4ff3 100644
--- a/drivers/spi/spi-rb4xx.c
+++ b/drivers/spi/spi-rb4xx.c
@@ -14,6 +14,7 @@
 #include <linux/platform_device.h>
 #include <linux/clk.h>
 #include <linux/spi/spi.h>
+#include <linux/of.h>
 
 #include <asm/mach-ath79/ar71xx_regs.h>
 
@@ -150,6 +151,7 @@ static int rb4xx_spi_probe(struct platform_device *pdev)
 	if (IS_ERR(ahb_clk))
 		return PTR_ERR(ahb_clk);
 
+	master->dev.of_node = pdev->dev.of_node;
 	master->bus_num = 0;
 	master->num_chipselect = 3;
 	master->mode_bits = SPI_TX_DUAL;
@@ -158,6 +160,11 @@ static int rb4xx_spi_probe(struct platform_device *pdev)
 	master->transfer_one = rb4xx_transfer_one;
 	master->set_cs = rb4xx_set_cs;
 
+	rbspi = spi_master_get_devdata(master);
+	rbspi->base = spi_base;
+	rbspi->clk = ahb_clk;
+	platform_set_drvdata(pdev, rbspi);
+
 	err = devm_spi_register_master(&pdev->dev, master);
 	if (err) {
 		dev_err(&pdev->dev, "failed to register SPI master\n");
@@ -168,11 +175,6 @@ static int rb4xx_spi_probe(struct platform_device *pdev)
 	if (err)
 		return err;
 
-	rbspi = spi_master_get_devdata(master);
-	rbspi->base = spi_base;
-	rbspi->clk = ahb_clk;
-	platform_set_drvdata(pdev, rbspi);
-
 	/* Enable SPI */
 	rb4xx_write(rbspi, AR71XX_SPI_REG_FS, AR71XX_SPI_FS_GPIO);
 
@@ -188,11 +190,18 @@ static int rb4xx_spi_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id rb4xx_spi_dt_match[] = {
+	{ .compatible = "mikrotik,rb4xx-spi" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, rb4xx_spi_dt_match);
+
 static struct platform_driver rb4xx_spi_drv = {
 	.probe = rb4xx_spi_probe,
 	.remove = rb4xx_spi_remove,
 	.driver = {
 		.name = "rb4xx-spi",
+		.of_match_table = of_match_ptr(rb4xx_spi_dt_match),
 	},
 };
 
diff --git a/drivers/spi/spi-rockchip.c b/drivers/spi/spi-rockchip.c
index 70ef63e0b6b8..9b8a5e1233c0 100644
--- a/drivers/spi/spi-rockchip.c
+++ b/drivers/spi/spi-rockchip.c
@@ -183,6 +183,8 @@ struct rockchip_spi {
 	u8 rsd;
 
 	bool cs_asserted[ROCKCHIP_SPI_MAX_CS_NUM];
+
+	bool slave_abort;
 };
 
 static inline void spi_enable_chip(struct rockchip_spi *rs, bool enable)
@@ -219,8 +221,8 @@ static u32 get_fifo_len(struct rockchip_spi *rs)
 
 static void rockchip_spi_set_cs(struct spi_device *spi, bool enable)
 {
-	struct spi_master *master = spi->master;
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = spi->controller;
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 	bool cs_asserted = !enable;
 
 	/* Return immediately for no-op */
@@ -244,10 +246,10 @@ static void rockchip_spi_set_cs(struct spi_device *spi, bool enable)
 	rs->cs_asserted[spi->chip_select] = cs_asserted;
 }
 
-static void rockchip_spi_handle_err(struct spi_master *master,
+static void rockchip_spi_handle_err(struct spi_controller *ctlr,
 				    struct spi_message *msg)
 {
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	/* stop running spi transfer
 	 * this also flushes both rx and tx fifos
@@ -258,10 +260,10 @@ static void rockchip_spi_handle_err(struct spi_master *master,
 	writel_relaxed(0, rs->regs + ROCKCHIP_SPI_IMR);
 
 	if (atomic_read(&rs->state) & TXDMA)
-		dmaengine_terminate_async(master->dma_tx);
+		dmaengine_terminate_async(ctlr->dma_tx);
 
 	if (atomic_read(&rs->state) & RXDMA)
-		dmaengine_terminate_async(master->dma_rx);
+		dmaengine_terminate_async(ctlr->dma_rx);
 }
 
 static void rockchip_spi_pio_writer(struct rockchip_spi *rs)
@@ -319,8 +321,8 @@ static void rockchip_spi_pio_reader(struct rockchip_spi *rs)
 
 static irqreturn_t rockchip_spi_isr(int irq, void *dev_id)
 {
-	struct spi_master *master = dev_id;
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = dev_id;
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	if (rs->tx_left)
 		rockchip_spi_pio_writer(rs);
@@ -329,7 +331,7 @@ static irqreturn_t rockchip_spi_isr(int irq, void *dev_id)
 	if (!rs->rx_left) {
 		spi_enable_chip(rs, false);
 		writel_relaxed(0, rs->regs + ROCKCHIP_SPI_IMR);
-		spi_finalize_current_transfer(master);
+		spi_finalize_current_transfer(ctlr);
 	}
 
 	return IRQ_HANDLED;
@@ -355,35 +357,35 @@ static int rockchip_spi_prepare_irq(struct rockchip_spi *rs,
 
 static void rockchip_spi_dma_rxcb(void *data)
 {
-	struct spi_master *master = data;
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = data;
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 	int state = atomic_fetch_andnot(RXDMA, &rs->state);
 
-	if (state & TXDMA)
+	if (state & TXDMA && !rs->slave_abort)
 		return;
 
 	spi_enable_chip(rs, false);
-	spi_finalize_current_transfer(master);
+	spi_finalize_current_transfer(ctlr);
 }
 
 static void rockchip_spi_dma_txcb(void *data)
 {
-	struct spi_master *master = data;
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = data;
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 	int state = atomic_fetch_andnot(TXDMA, &rs->state);
 
-	if (state & RXDMA)
+	if (state & RXDMA && !rs->slave_abort)
 		return;
 
 	/* Wait until the FIFO data completely. */
 	wait_for_idle(rs);
 
 	spi_enable_chip(rs, false);
-	spi_finalize_current_transfer(master);
+	spi_finalize_current_transfer(ctlr);
 }
 
 static int rockchip_spi_prepare_dma(struct rockchip_spi *rs,
-		struct spi_master *master, struct spi_transfer *xfer)
+		struct spi_controller *ctlr, struct spi_transfer *xfer)
 {
 	struct dma_async_tx_descriptor *rxdesc, *txdesc;
 
@@ -398,17 +400,17 @@ static int rockchip_spi_prepare_dma(struct rockchip_spi *rs,
 			.src_maxburst = 1,
 		};
 
-		dmaengine_slave_config(master->dma_rx, &rxconf);
+		dmaengine_slave_config(ctlr->dma_rx, &rxconf);
 
 		rxdesc = dmaengine_prep_slave_sg(
-				master->dma_rx,
+				ctlr->dma_rx,
 				xfer->rx_sg.sgl, xfer->rx_sg.nents,
 				DMA_DEV_TO_MEM, DMA_PREP_INTERRUPT);
 		if (!rxdesc)
 			return -EINVAL;
 
 		rxdesc->callback = rockchip_spi_dma_rxcb;
-		rxdesc->callback_param = master;
+		rxdesc->callback_param = ctlr;
 	}
 
 	txdesc = NULL;
@@ -420,27 +422,27 @@ static int rockchip_spi_prepare_dma(struct rockchip_spi *rs,
 			.dst_maxburst = rs->fifo_len / 4,
 		};
 
-		dmaengine_slave_config(master->dma_tx, &txconf);
+		dmaengine_slave_config(ctlr->dma_tx, &txconf);
 
 		txdesc = dmaengine_prep_slave_sg(
-				master->dma_tx,
+				ctlr->dma_tx,
 				xfer->tx_sg.sgl, xfer->tx_sg.nents,
 				DMA_MEM_TO_DEV, DMA_PREP_INTERRUPT);
 		if (!txdesc) {
 			if (rxdesc)
-				dmaengine_terminate_sync(master->dma_rx);
+				dmaengine_terminate_sync(ctlr->dma_rx);
 			return -EINVAL;
 		}
 
 		txdesc->callback = rockchip_spi_dma_txcb;
-		txdesc->callback_param = master;
+		txdesc->callback_param = ctlr;
 	}
 
 	/* rx must be started before tx due to spi instinct */
 	if (rxdesc) {
 		atomic_or(RXDMA, &rs->state);
 		dmaengine_submit(rxdesc);
-		dma_async_issue_pending(master->dma_rx);
+		dma_async_issue_pending(ctlr->dma_rx);
 	}
 
 	spi_enable_chip(rs, true);
@@ -448,7 +450,7 @@ static int rockchip_spi_prepare_dma(struct rockchip_spi *rs,
 	if (txdesc) {
 		atomic_or(TXDMA, &rs->state);
 		dmaengine_submit(txdesc);
-		dma_async_issue_pending(master->dma_tx);
+		dma_async_issue_pending(ctlr->dma_tx);
 	}
 
 	/* 1 means the transfer is in progress */
@@ -457,7 +459,7 @@ static int rockchip_spi_prepare_dma(struct rockchip_spi *rs,
 
 static void rockchip_spi_config(struct rockchip_spi *rs,
 		struct spi_device *spi, struct spi_transfer *xfer,
-		bool use_dma)
+		bool use_dma, bool slave_mode)
 {
 	u32 cr0 = CR0_FRF_SPI  << CR0_FRF_OFFSET
 	        | CR0_BHT_8BIT << CR0_BHT_OFFSET
@@ -466,6 +468,10 @@ static void rockchip_spi_config(struct rockchip_spi *rs,
 	u32 cr1;
 	u32 dmacr = 0;
 
+	if (slave_mode)
+		cr0 |= CR0_OPM_SLAVE << CR0_OPM_OFFSET;
+	rs->slave_abort = false;
+
 	cr0 |= rs->rsd << CR0_RSD_OFFSET;
 	cr0 |= (spi->mode & 0x3U) << CR0_SCPH_OFFSET;
 	if (spi->mode & SPI_LSB_FIRST)
@@ -493,7 +499,7 @@ static void rockchip_spi_config(struct rockchip_spi *rs,
 		break;
 	default:
 		/* we only whitelist 4, 8 and 16 bit words in
-		 * master->bits_per_word_mask, so this shouldn't
+		 * ctlr->bits_per_word_mask, so this shouldn't
 		 * happen
 		 */
 		unreachable();
@@ -535,12 +541,22 @@ static size_t rockchip_spi_max_transfer_size(struct spi_device *spi)
 	return ROCKCHIP_SPI_MAX_TRANLEN;
 }
 
+static int rockchip_spi_slave_abort(struct spi_controller *ctlr)
+{
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
+
+	rs->slave_abort = true;
+	complete(&ctlr->xfer_completion);
+
+	return 0;
+}
+
 static int rockchip_spi_transfer_one(
-		struct spi_master *master,
+		struct spi_controller *ctlr,
 		struct spi_device *spi,
 		struct spi_transfer *xfer)
 {
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 	bool use_dma;
 
 	WARN_ON(readl_relaxed(rs->regs + ROCKCHIP_SPI_SSIENR) &&
@@ -558,21 +574,21 @@ static int rockchip_spi_transfer_one(
 
 	rs->n_bytes = xfer->bits_per_word <= 8 ? 1 : 2;
 
-	use_dma = master->can_dma ? master->can_dma(master, spi, xfer) : false;
+	use_dma = ctlr->can_dma ? ctlr->can_dma(ctlr, spi, xfer) : false;
 
-	rockchip_spi_config(rs, spi, xfer, use_dma);
+	rockchip_spi_config(rs, spi, xfer, use_dma, ctlr->slave);
 
 	if (use_dma)
-		return rockchip_spi_prepare_dma(rs, master, xfer);
+		return rockchip_spi_prepare_dma(rs, ctlr, xfer);
 
 	return rockchip_spi_prepare_irq(rs, xfer);
 }
 
-static bool rockchip_spi_can_dma(struct spi_master *master,
+static bool rockchip_spi_can_dma(struct spi_controller *ctlr,
 				 struct spi_device *spi,
 				 struct spi_transfer *xfer)
 {
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 	unsigned int bytes_per_word = xfer->bits_per_word <= 8 ? 1 : 2;
 
 	/* if the numbor of spi words to transfer is less than the fifo
@@ -586,44 +602,55 @@ static int rockchip_spi_probe(struct platform_device *pdev)
 {
 	int ret;
 	struct rockchip_spi *rs;
-	struct spi_master *master;
+	struct spi_controller *ctlr;
 	struct resource *mem;
+	struct device_node *np = pdev->dev.of_node;
 	u32 rsd_nsecs;
+	bool slave_mode;
+
+	slave_mode = of_property_read_bool(np, "spi-slave");
+
+	if (slave_mode)
+		ctlr = spi_alloc_slave(&pdev->dev,
+				sizeof(struct rockchip_spi));
+	else
+		ctlr = spi_alloc_master(&pdev->dev,
+				sizeof(struct rockchip_spi));
 
-	master = spi_alloc_master(&pdev->dev, sizeof(struct rockchip_spi));
-	if (!master)
+	if (!ctlr)
 		return -ENOMEM;
 
-	platform_set_drvdata(pdev, master);
+	platform_set_drvdata(pdev, ctlr);
 
-	rs = spi_master_get_devdata(master);
+	rs = spi_controller_get_devdata(ctlr);
+	ctlr->slave = slave_mode;
 
 	/* Get basic io resource and map it */
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	rs->regs = devm_ioremap_resource(&pdev->dev, mem);
 	if (IS_ERR(rs->regs)) {
 		ret =  PTR_ERR(rs->regs);
-		goto err_put_master;
+		goto err_put_ctlr;
 	}
 
 	rs->apb_pclk = devm_clk_get(&pdev->dev, "apb_pclk");
 	if (IS_ERR(rs->apb_pclk)) {
 		dev_err(&pdev->dev, "Failed to get apb_pclk\n");
 		ret = PTR_ERR(rs->apb_pclk);
-		goto err_put_master;
+		goto err_put_ctlr;
 	}
 
 	rs->spiclk = devm_clk_get(&pdev->dev, "spiclk");
 	if (IS_ERR(rs->spiclk)) {
 		dev_err(&pdev->dev, "Failed to get spi_pclk\n");
 		ret = PTR_ERR(rs->spiclk);
-		goto err_put_master;
+		goto err_put_ctlr;
 	}
 
 	ret = clk_prepare_enable(rs->apb_pclk);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "Failed to enable apb_pclk\n");
-		goto err_put_master;
+		goto err_put_ctlr;
 	}
 
 	ret = clk_prepare_enable(rs->spiclk);
@@ -639,7 +666,7 @@ static int rockchip_spi_probe(struct platform_device *pdev)
 		goto err_disable_spiclk;
 
 	ret = devm_request_threaded_irq(&pdev->dev, ret, rockchip_spi_isr, NULL,
-			IRQF_ONESHOT, dev_name(&pdev->dev), master);
+			IRQF_ONESHOT, dev_name(&pdev->dev), ctlr);
 	if (ret)
 		goto err_disable_spiclk;
 
@@ -673,78 +700,90 @@ static int rockchip_spi_probe(struct platform_device *pdev)
 	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
 
-	master->auto_runtime_pm = true;
-	master->bus_num = pdev->id;
-	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_LOOP | SPI_LSB_FIRST;
-	master->num_chipselect = ROCKCHIP_SPI_MAX_CS_NUM;
-	master->dev.of_node = pdev->dev.of_node;
-	master->bits_per_word_mask = SPI_BPW_MASK(16) | SPI_BPW_MASK(8) | SPI_BPW_MASK(4);
-	master->min_speed_hz = rs->freq / BAUDR_SCKDV_MAX;
-	master->max_speed_hz = min(rs->freq / BAUDR_SCKDV_MIN, MAX_SCLK_OUT);
-
-	master->set_cs = rockchip_spi_set_cs;
-	master->transfer_one = rockchip_spi_transfer_one;
-	master->max_transfer_size = rockchip_spi_max_transfer_size;
-	master->handle_err = rockchip_spi_handle_err;
-	master->flags = SPI_MASTER_GPIO_SS;
-
-	master->dma_tx = dma_request_chan(rs->dev, "tx");
-	if (IS_ERR(master->dma_tx)) {
+	ctlr->auto_runtime_pm = true;
+	ctlr->bus_num = pdev->id;
+	ctlr->mode_bits = SPI_CPOL | SPI_CPHA | SPI_LOOP | SPI_LSB_FIRST;
+	if (slave_mode) {
+		ctlr->mode_bits |= SPI_NO_CS;
+		ctlr->slave_abort = rockchip_spi_slave_abort;
+	} else {
+		ctlr->flags = SPI_MASTER_GPIO_SS;
+		ctlr->max_native_cs = ROCKCHIP_SPI_MAX_CS_NUM;
+		/*
+		 * rk spi0 has two native cs, spi1..5 one cs only
+		 * if num-cs is missing in the dts, default to 1
+		 */
+		if (of_property_read_u16(np, "num-cs", &ctlr->num_chipselect))
+			ctlr->num_chipselect = 1;
+		ctlr->use_gpio_descriptors = true;
+	}
+	ctlr->dev.of_node = pdev->dev.of_node;
+	ctlr->bits_per_word_mask = SPI_BPW_MASK(16) | SPI_BPW_MASK(8) | SPI_BPW_MASK(4);
+	ctlr->min_speed_hz = rs->freq / BAUDR_SCKDV_MAX;
+	ctlr->max_speed_hz = min(rs->freq / BAUDR_SCKDV_MIN, MAX_SCLK_OUT);
+
+	ctlr->set_cs = rockchip_spi_set_cs;
+	ctlr->transfer_one = rockchip_spi_transfer_one;
+	ctlr->max_transfer_size = rockchip_spi_max_transfer_size;
+	ctlr->handle_err = rockchip_spi_handle_err;
+
+	ctlr->dma_tx = dma_request_chan(rs->dev, "tx");
+	if (IS_ERR(ctlr->dma_tx)) {
 		/* Check tx to see if we need defer probing driver */
-		if (PTR_ERR(master->dma_tx) == -EPROBE_DEFER) {
+		if (PTR_ERR(ctlr->dma_tx) == -EPROBE_DEFER) {
 			ret = -EPROBE_DEFER;
 			goto err_disable_pm_runtime;
 		}
 		dev_warn(rs->dev, "Failed to request TX DMA channel\n");
-		master->dma_tx = NULL;
+		ctlr->dma_tx = NULL;
 	}
 
-	master->dma_rx = dma_request_chan(rs->dev, "rx");
-	if (IS_ERR(master->dma_rx)) {
-		if (PTR_ERR(master->dma_rx) == -EPROBE_DEFER) {
+	ctlr->dma_rx = dma_request_chan(rs->dev, "rx");
+	if (IS_ERR(ctlr->dma_rx)) {
+		if (PTR_ERR(ctlr->dma_rx) == -EPROBE_DEFER) {
 			ret = -EPROBE_DEFER;
 			goto err_free_dma_tx;
 		}
 		dev_warn(rs->dev, "Failed to request RX DMA channel\n");
-		master->dma_rx = NULL;
+		ctlr->dma_rx = NULL;
 	}
 
-	if (master->dma_tx && master->dma_rx) {
+	if (ctlr->dma_tx && ctlr->dma_rx) {
 		rs->dma_addr_tx = mem->start + ROCKCHIP_SPI_TXDR;
 		rs->dma_addr_rx = mem->start + ROCKCHIP_SPI_RXDR;
-		master->can_dma = rockchip_spi_can_dma;
+		ctlr->can_dma = rockchip_spi_can_dma;
 	}
 
-	ret = devm_spi_register_master(&pdev->dev, master);
+	ret = devm_spi_register_controller(&pdev->dev, ctlr);
 	if (ret < 0) {
-		dev_err(&pdev->dev, "Failed to register master\n");
+		dev_err(&pdev->dev, "Failed to register controller\n");
 		goto err_free_dma_rx;
 	}
 
 	return 0;
 
 err_free_dma_rx:
-	if (master->dma_rx)
-		dma_release_channel(master->dma_rx);
+	if (ctlr->dma_rx)
+		dma_release_channel(ctlr->dma_rx);
 err_free_dma_tx:
-	if (master->dma_tx)
-		dma_release_channel(master->dma_tx);
+	if (ctlr->dma_tx)
+		dma_release_channel(ctlr->dma_tx);
 err_disable_pm_runtime:
 	pm_runtime_disable(&pdev->dev);
 err_disable_spiclk:
 	clk_disable_unprepare(rs->spiclk);
 err_disable_apbclk:
 	clk_disable_unprepare(rs->apb_pclk);
-err_put_master:
-	spi_master_put(master);
+err_put_ctlr:
+	spi_controller_put(ctlr);
 
 	return ret;
 }
 
 static int rockchip_spi_remove(struct platform_device *pdev)
 {
-	struct spi_master *master = spi_master_get(platform_get_drvdata(pdev));
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = spi_controller_get(platform_get_drvdata(pdev));
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	pm_runtime_get_sync(&pdev->dev);
 
@@ -755,12 +794,12 @@ static int rockchip_spi_remove(struct platform_device *pdev)
 	pm_runtime_disable(&pdev->dev);
 	pm_runtime_set_suspended(&pdev->dev);
 
-	if (master->dma_tx)
-		dma_release_channel(master->dma_tx);
-	if (master->dma_rx)
-		dma_release_channel(master->dma_rx);
+	if (ctlr->dma_tx)
+		dma_release_channel(ctlr->dma_tx);
+	if (ctlr->dma_rx)
+		dma_release_channel(ctlr->dma_rx);
 
-	spi_master_put(master);
+	spi_controller_put(ctlr);
 
 	return 0;
 }
@@ -769,9 +808,9 @@ static int rockchip_spi_remove(struct platform_device *pdev)
 static int rockchip_spi_suspend(struct device *dev)
 {
 	int ret;
-	struct spi_master *master = dev_get_drvdata(dev);
+	struct spi_controller *ctlr = dev_get_drvdata(dev);
 
-	ret = spi_master_suspend(master);
+	ret = spi_controller_suspend(ctlr);
 	if (ret < 0)
 		return ret;
 
@@ -787,8 +826,8 @@ static int rockchip_spi_suspend(struct device *dev)
 static int rockchip_spi_resume(struct device *dev)
 {
 	int ret;
-	struct spi_master *master = dev_get_drvdata(dev);
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = dev_get_drvdata(dev);
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	pinctrl_pm_select_default_state(dev);
 
@@ -796,7 +835,7 @@ static int rockchip_spi_resume(struct device *dev)
 	if (ret < 0)
 		return ret;
 
-	ret = spi_master_resume(master);
+	ret = spi_controller_resume(ctlr);
 	if (ret < 0) {
 		clk_disable_unprepare(rs->spiclk);
 		clk_disable_unprepare(rs->apb_pclk);
@@ -809,8 +848,8 @@ static int rockchip_spi_resume(struct device *dev)
 #ifdef CONFIG_PM
 static int rockchip_spi_runtime_suspend(struct device *dev)
 {
-	struct spi_master *master = dev_get_drvdata(dev);
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = dev_get_drvdata(dev);
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	clk_disable_unprepare(rs->spiclk);
 	clk_disable_unprepare(rs->apb_pclk);
@@ -821,8 +860,8 @@ static int rockchip_spi_runtime_suspend(struct device *dev)
 static int rockchip_spi_runtime_resume(struct device *dev)
 {
 	int ret;
-	struct spi_master *master = dev_get_drvdata(dev);
-	struct rockchip_spi *rs = spi_master_get_devdata(master);
+	struct spi_controller *ctlr = dev_get_drvdata(dev);
+	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	ret = clk_prepare_enable(rs->apb_pclk);
 	if (ret < 0)
diff --git a/drivers/spi/spi-sc18is602.c b/drivers/spi/spi-sc18is602.c
index 5497eeb3bf3e..ee0f3edf49cd 100644
--- a/drivers/spi/spi-sc18is602.c
+++ b/drivers/spi/spi-sc18is602.c
@@ -345,6 +345,6 @@ static struct i2c_driver sc18is602_driver = {
 
 module_i2c_driver(sc18is602_driver);
 
-MODULE_DESCRIPTION("SC18IC602/603 SPI Master Driver");
+MODULE_DESCRIPTION("SC18IS602/603 SPI Master Driver");
 MODULE_AUTHOR("Guenter Roeck");
 MODULE_LICENSE("GPL");
diff --git a/drivers/spi/spi-sh-msiof.c b/drivers/spi/spi-sh-msiof.c
index 1c11a00a2c36..b2579af0e3eb 100644
--- a/drivers/spi/spi-sh-msiof.c
+++ b/drivers/spi/spi-sh-msiof.c
@@ -1398,7 +1398,7 @@ static int sh_msiof_spi_resume(struct device *dev)
 
 static SIMPLE_DEV_PM_OPS(sh_msiof_spi_pm_ops, sh_msiof_spi_suspend,
 			 sh_msiof_spi_resume);
-#define DEV_PM_OPS	&sh_msiof_spi_pm_ops
+#define DEV_PM_OPS	(&sh_msiof_spi_pm_ops)
 #else
 #define DEV_PM_OPS	NULL
 #endif /* CONFIG_PM_SLEEP */
diff --git a/drivers/spi/spi-sprd-adi.c b/drivers/spi/spi-sprd-adi.c
index 87dadb6b8ebf..88e6543648cb 100644
--- a/drivers/spi/spi-sprd-adi.c
+++ b/drivers/spi/spi-sprd-adi.c
@@ -319,7 +319,7 @@ static int sprd_adi_transfer_one(struct spi_controller *ctlr,
 
 static void sprd_adi_set_wdt_rst_mode(struct sprd_adi *sadi)
 {
-#ifdef CONFIG_SPRD_WATCHDOG
+#if IS_ENABLED(CONFIG_SPRD_WATCHDOG)
 	u32 val;
 
 	/* Set default watchdog reboot mode */
diff --git a/drivers/spi/spi-stm32-qspi.c b/drivers/spi/spi-stm32-qspi.c
index d066f5144c3e..3c44bb2fd9b1 100644
--- a/drivers/spi/spi-stm32-qspi.c
+++ b/drivers/spi/spi-stm32-qspi.c
@@ -16,6 +16,7 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/pinctrl/consumer.h>
+#include <linux/pm_runtime.h>
 #include <linux/platform_device.h>
 #include <linux/reset.h>
 #include <linux/sizes.h>
@@ -87,6 +88,7 @@
 #define STM32_BUSY_TIMEOUT_US 100000
 #define STM32_ABT_TIMEOUT_US 100000
 #define STM32_COMP_TIMEOUT_MS 1000
+#define STM32_AUTOSUSPEND_DELAY -1
 
 struct stm32_qspi_flash {
 	struct stm32_qspi *qspi;
@@ -431,10 +433,17 @@ static int stm32_qspi_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
 	struct stm32_qspi *qspi = spi_controller_get_devdata(mem->spi->master);
 	int ret;
 
+	ret = pm_runtime_get_sync(qspi->dev);
+	if (ret < 0)
+		return ret;
+
 	mutex_lock(&qspi->lock);
 	ret = stm32_qspi_send(mem, op);
 	mutex_unlock(&qspi->lock);
 
+	pm_runtime_mark_last_busy(qspi->dev);
+	pm_runtime_put_autosuspend(qspi->dev);
+
 	return ret;
 }
 
@@ -444,6 +453,7 @@ static int stm32_qspi_setup(struct spi_device *spi)
 	struct stm32_qspi *qspi = spi_controller_get_devdata(ctrl);
 	struct stm32_qspi_flash *flash;
 	u32 presc;
+	int ret;
 
 	if (ctrl->busy)
 		return -EBUSY;
@@ -451,6 +461,10 @@ static int stm32_qspi_setup(struct spi_device *spi)
 	if (!spi->max_speed_hz)
 		return -EINVAL;
 
+	ret = pm_runtime_get_sync(qspi->dev);
+	if (ret < 0)
+		return ret;
+
 	presc = DIV_ROUND_UP(qspi->clk_rate, spi->max_speed_hz) - 1;
 
 	flash = &qspi->flash[spi->chip_select];
@@ -467,6 +481,9 @@ static int stm32_qspi_setup(struct spi_device *spi)
 	writel_relaxed(qspi->dcr_reg, qspi->io_base + QSPI_DCR);
 	mutex_unlock(&qspi->lock);
 
+	pm_runtime_mark_last_busy(qspi->dev);
+	pm_runtime_put_autosuspend(qspi->dev);
+
 	return 0;
 }
 
@@ -538,10 +555,15 @@ static const struct spi_controller_mem_ops stm32_qspi_mem_ops = {
 
 static void stm32_qspi_release(struct stm32_qspi *qspi)
 {
+	pm_runtime_get_sync(qspi->dev);
 	/* disable qspi */
 	writel_relaxed(0, qspi->io_base + QSPI_CR);
 	stm32_qspi_dma_free(qspi);
 	mutex_destroy(&qspi->lock);
+	pm_runtime_put_noidle(qspi->dev);
+	pm_runtime_disable(qspi->dev);
+	pm_runtime_set_suspended(qspi->dev);
+	pm_runtime_dont_use_autosuspend(qspi->dev);
 	clk_disable_unprepare(qspi->clk);
 }
 
@@ -643,9 +665,20 @@ static int stm32_qspi_probe(struct platform_device *pdev)
 	ctrl->num_chipselect = STM32_QSPI_MAX_NORCHIP;
 	ctrl->dev.of_node = dev->of_node;
 
+	pm_runtime_set_autosuspend_delay(dev, STM32_AUTOSUSPEND_DELAY);
+	pm_runtime_use_autosuspend(dev);
+	pm_runtime_set_active(dev);
+	pm_runtime_enable(dev);
+	pm_runtime_get_noresume(dev);
+
 	ret = devm_spi_register_master(dev, ctrl);
-	if (!ret)
-		return 0;
+	if (ret)
+		goto err_qspi_release;
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
+
+	return 0;
 
 err_qspi_release:
 	stm32_qspi_release(qspi);
@@ -660,14 +693,28 @@ static int stm32_qspi_remove(struct platform_device *pdev)
 	struct stm32_qspi *qspi = platform_get_drvdata(pdev);
 
 	stm32_qspi_release(qspi);
+
 	return 0;
 }
 
-static int __maybe_unused stm32_qspi_suspend(struct device *dev)
+static int __maybe_unused stm32_qspi_runtime_suspend(struct device *dev)
 {
 	struct stm32_qspi *qspi = dev_get_drvdata(dev);
 
 	clk_disable_unprepare(qspi->clk);
+
+	return 0;
+}
+
+static int __maybe_unused stm32_qspi_runtime_resume(struct device *dev)
+{
+	struct stm32_qspi *qspi = dev_get_drvdata(dev);
+
+	return clk_prepare_enable(qspi->clk);
+}
+
+static int __maybe_unused stm32_qspi_suspend(struct device *dev)
+{
 	pinctrl_pm_select_sleep_state(dev);
 
 	return 0;
@@ -683,10 +730,17 @@ static int __maybe_unused stm32_qspi_resume(struct device *dev)
 	writel_relaxed(qspi->cr_reg, qspi->io_base + QSPI_CR);
 	writel_relaxed(qspi->dcr_reg, qspi->io_base + QSPI_DCR);
 
+	pm_runtime_mark_last_busy(qspi->dev);
+	pm_runtime_put_autosuspend(qspi->dev);
+
 	return 0;
 }
 
-static SIMPLE_DEV_PM_OPS(stm32_qspi_pm_ops, stm32_qspi_suspend, stm32_qspi_resume);
+static const struct dev_pm_ops stm32_qspi_pm_ops = {
+	SET_RUNTIME_PM_OPS(stm32_qspi_runtime_suspend,
+			   stm32_qspi_runtime_resume, NULL)
+	SET_SYSTEM_SLEEP_PM_OPS(stm32_qspi_suspend, stm32_qspi_resume)
+};
 
 static const struct of_device_id stm32_qspi_match[] = {
 	{.compatible = "st,stm32f469-qspi"},
diff --git a/drivers/spi/spi-stm32.c b/drivers/spi/spi-stm32.c
index 44ac6eb3298d..4c643dfc7fbb 100644
--- a/drivers/spi/spi-stm32.c
+++ b/drivers/spi/spi-stm32.c
@@ -811,7 +811,9 @@ static irqreturn_t stm32f4_spi_irq_event(int irq, void *dev_id)
 		mask |= STM32F4_SPI_SR_TXE;
 	}
 
-	if (!spi->cur_usedma && spi->cur_comm == SPI_FULL_DUPLEX) {
+	if (!spi->cur_usedma && (spi->cur_comm == SPI_FULL_DUPLEX ||
+				spi->cur_comm == SPI_SIMPLEX_RX ||
+				spi->cur_comm == SPI_3WIRE_RX)) {
 		/* TXE flag is set and is handled when RXNE flag occurs */
 		sr &= ~STM32F4_SPI_SR_TXE;
 		mask |= STM32F4_SPI_SR_RXNE | STM32F4_SPI_SR_OVR;
@@ -850,7 +852,7 @@ static irqreturn_t stm32f4_spi_irq_event(int irq, void *dev_id)
 		stm32f4_spi_read_rx(spi);
 		if (spi->rx_len == 0)
 			end = true;
-		else /* Load data for discontinuous mode */
+		else if (spi->tx_buf)/* Load data for discontinuous mode */
 			stm32f4_spi_write_tx(spi);
 	}
 
@@ -1151,7 +1153,9 @@ static int stm32f4_spi_transfer_one_irq(struct stm32_spi *spi)
 	/* Enable the interrupts relative to the current communication mode */
 	if (spi->cur_comm == SPI_SIMPLEX_TX || spi->cur_comm == SPI_3WIRE_TX) {
 		cr2 |= STM32F4_SPI_CR2_TXEIE;
-	} else if (spi->cur_comm == SPI_FULL_DUPLEX) {
+	} else if (spi->cur_comm == SPI_FULL_DUPLEX ||
+				spi->cur_comm == SPI_SIMPLEX_RX ||
+				spi->cur_comm == SPI_3WIRE_RX) {
 		/* In transmit-only mode, the OVR flag is set in the SR register
 		 * since the received data are never read. Therefore set OVR
 		 * interrupt only when rx buffer is available.
@@ -1462,10 +1466,16 @@ static int stm32f4_spi_set_mode(struct stm32_spi *spi, unsigned int comm_type)
 		stm32_spi_set_bits(spi, STM32F4_SPI_CR1,
 					STM32F4_SPI_CR1_BIDIMODE |
 					STM32F4_SPI_CR1_BIDIOE);
-	} else if (comm_type == SPI_FULL_DUPLEX) {
+	} else if (comm_type == SPI_FULL_DUPLEX ||
+				comm_type == SPI_SIMPLEX_RX) {
 		stm32_spi_clr_bits(spi, STM32F4_SPI_CR1,
 					STM32F4_SPI_CR1_BIDIMODE |
 					STM32F4_SPI_CR1_BIDIOE);
+	} else if (comm_type == SPI_3WIRE_RX) {
+		stm32_spi_set_bits(spi, STM32F4_SPI_CR1,
+					STM32F4_SPI_CR1_BIDIMODE);
+		stm32_spi_clr_bits(spi, STM32F4_SPI_CR1,
+					STM32F4_SPI_CR1_BIDIOE);
 	} else {
 		return -EINVAL;
 	}
@@ -1906,6 +1916,7 @@ static int stm32_spi_probe(struct platform_device *pdev)
 	master->prepare_message = stm32_spi_prepare_msg;
 	master->transfer_one = stm32_spi_transfer_one;
 	master->unprepare_message = stm32_spi_unprepare_msg;
+	master->flags = SPI_MASTER_MUST_TX;
 
 	spi->dma_tx = dma_request_chan(spi->dev, "tx");
 	if (IS_ERR(spi->dma_tx)) {
diff --git a/drivers/spi/spi-sun6i.c b/drivers/spi/spi-sun6i.c
index ec7967be9e2f..ecea15534c42 100644
--- a/drivers/spi/spi-sun6i.c
+++ b/drivers/spi/spi-sun6i.c
@@ -470,6 +470,7 @@ static int sun6i_spi_probe(struct platform_device *pdev)
 
 	master->max_speed_hz = 100 * 1000 * 1000;
 	master->min_speed_hz = 3 * 1000;
+	master->use_gpio_descriptors = true;
 	master->set_cs = sun6i_spi_set_cs;
 	master->transfer_one = sun6i_spi_transfer_one;
 	master->num_chipselect = 4;
diff --git a/drivers/spi/spi-tegra114.c b/drivers/spi/spi-tegra114.c
index 83edabdb41ad..c2c58871a947 100644
--- a/drivers/spi/spi-tegra114.c
+++ b/drivers/spi/spi-tegra114.c
@@ -1398,6 +1398,7 @@ static int tegra_spi_probe(struct platform_device *pdev)
 	ret = pm_runtime_get_sync(&pdev->dev);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "pm runtime get failed, e = %d\n", ret);
+		pm_runtime_put_noidle(&pdev->dev);
 		goto exit_pm_disable;
 	}
 
diff --git a/drivers/spi/spi-tegra20-sflash.c b/drivers/spi/spi-tegra20-sflash.c
index 514429379206..02cf5f463ba6 100644
--- a/drivers/spi/spi-tegra20-sflash.c
+++ b/drivers/spi/spi-tegra20-sflash.c
@@ -491,6 +491,7 @@ static int tegra_sflash_probe(struct platform_device *pdev)
 	ret = pm_runtime_get_sync(&pdev->dev);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "pm runtime get failed, e = %d\n", ret);
+		pm_runtime_put_noidle(&pdev->dev);
 		goto exit_pm_disable;
 	}
 
diff --git a/drivers/spi/spi-tegra20-slink.c b/drivers/spi/spi-tegra20-slink.c
index 7f4d932dade7..a07b72e9c344 100644
--- a/drivers/spi/spi-tegra20-slink.c
+++ b/drivers/spi/spi-tegra20-slink.c
@@ -1118,6 +1118,7 @@ static int tegra_slink_probe(struct platform_device *pdev)
 	ret = pm_runtime_get_sync(&pdev->dev);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "pm runtime get failed, e = %d\n", ret);
+		pm_runtime_put_noidle(&pdev->dev);
 		goto exit_pm_disable;
 	}
 	tspi->def_command_reg  = SLINK_M_S;
diff --git a/drivers/spi/spi-uniphier.c b/drivers/spi/spi-uniphier.c
index 0fa50979644d..6a9ef8ee3cc9 100644
--- a/drivers/spi/spi-uniphier.c
+++ b/drivers/spi/spi-uniphier.c
@@ -659,8 +659,7 @@ static int uniphier_spi_probe(struct platform_device *pdev)
 	priv->master = master;
 	priv->is_save_param = false;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->base = devm_ioremap_resource(&pdev->dev, res);
+	priv->base = devm_platform_get_and_ioremap_resource(pdev, 0, &res);
 	if (IS_ERR(priv->base)) {
 		ret = PTR_ERR(priv->base);
 		goto out_master_put;
@@ -716,8 +715,10 @@ static int uniphier_spi_probe(struct platform_device *pdev)
 
 	master->dma_tx = dma_request_chan(&pdev->dev, "tx");
 	if (IS_ERR_OR_NULL(master->dma_tx)) {
-		if (PTR_ERR(master->dma_tx) == -EPROBE_DEFER)
+		if (PTR_ERR(master->dma_tx) == -EPROBE_DEFER) {
+			ret = -EPROBE_DEFER;
 			goto out_disable_clk;
+		}
 		master->dma_tx = NULL;
 		dma_tx_burst = INT_MAX;
 	} else {
@@ -732,8 +733,10 @@ static int uniphier_spi_probe(struct platform_device *pdev)
 
 	master->dma_rx = dma_request_chan(&pdev->dev, "rx");
 	if (IS_ERR_OR_NULL(master->dma_rx)) {
-		if (PTR_ERR(master->dma_rx) == -EPROBE_DEFER)
+		if (PTR_ERR(master->dma_rx) == -EPROBE_DEFER) {
+			ret = -EPROBE_DEFER;
 			goto out_disable_clk;
+		}
 		master->dma_rx = NULL;
 		dma_rx_burst = INT_MAX;
 	} else {
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index c92c89467e7e..8158e281f354 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -1023,7 +1023,8 @@ static int spi_map_msg(struct spi_controller *ctlr, struct spi_message *msg)
 	void *tmp;
 	unsigned int max_tx, max_rx;
 
-	if (ctlr->flags & (SPI_CONTROLLER_MUST_RX | SPI_CONTROLLER_MUST_TX)) {
+	if ((ctlr->flags & (SPI_CONTROLLER_MUST_RX | SPI_CONTROLLER_MUST_TX))
+		&& !(msg->spi->mode & SPI_3WIRE)) {
 		max_tx = 0;
 		max_rx = 0;
 
@@ -1075,7 +1076,7 @@ static int spi_transfer_wait(struct spi_controller *ctlr,
 {
 	struct spi_statistics *statm = &ctlr->statistics;
 	struct spi_statistics *stats = &msg->spi->statistics;
-	unsigned long long ms = 1;
+	unsigned long long ms;
 
 	if (spi_controller_is_slave(ctlr)) {
 		if (wait_for_completion_interruptible(&ctlr->xfer_completion)) {
@@ -1160,6 +1161,8 @@ int spi_delay_exec(struct spi_delay *_delay, struct spi_transfer *xfer)
 {
 	int delay;
 
+	might_sleep();
+
 	if (!_delay)
 		return -EINVAL;
 
@@ -2111,6 +2114,7 @@ static int acpi_spi_add_resource(struct acpi_resource *ares, void *data)
 			}
 
 			lookup->max_speed_hz = sb->connection_speed;
+			lookup->bits_per_word = sb->data_bit_length;
 
 			if (sb->clock_phase == ACPI_SPI_SECOND_PHASE)
 				lookup->mode |= SPI_CPHA;
@@ -2760,6 +2764,8 @@ void spi_unregister_controller(struct spi_controller *ctlr)
 	struct spi_controller *found;
 	int id = ctlr->bus_num;
 
+	device_for_each_child(&ctlr->dev, NULL, __unregister);
+
 	/* First make sure that this controller was ever added */
 	mutex_lock(&board_lock);
 	found = idr_find(&spi_master_idr, id);
@@ -2772,7 +2778,6 @@ void spi_unregister_controller(struct spi_controller *ctlr)
 	list_del(&ctlr->list);
 	mutex_unlock(&board_lock);
 
-	device_for_each_child(&ctlr->dev, NULL, __unregister);
 	device_unregister(&ctlr->dev);
 	/* free bus id */
 	mutex_lock(&board_lock);
@@ -3853,8 +3858,7 @@ static u8	*buf;
  * is zero for success, else a negative errno status code.
  * This call may only be used from a context that may sleep.
  *
- * Parameters to this routine are always copied using a small buffer;
- * portable code should never use this for more than 32 bytes.
+ * Parameters to this routine are always copied using a small buffer.
  * Performance-sensitive or bulk transfer code should instead use
  * spi_{async,sync}() calls with dma-safe buffers.
  *
diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c
index 80dd1025b953..d753df700e9e 100644
--- a/drivers/spi/spidev.c
+++ b/drivers/spi/spidev.c
@@ -62,7 +62,8 @@ static DECLARE_BITMAP(minors, N_SPI_MINORS);
 #define SPI_MODE_MASK		(SPI_CPHA | SPI_CPOL | SPI_CS_HIGH \
 				| SPI_LSB_FIRST | SPI_3WIRE | SPI_LOOP \
 				| SPI_NO_CS | SPI_READY | SPI_TX_DUAL \
-				| SPI_TX_QUAD | SPI_RX_DUAL | SPI_RX_QUAD)
+				| SPI_TX_QUAD | SPI_TX_OCTAL | SPI_RX_DUAL \
+				| SPI_RX_QUAD | SPI_RX_OCTAL)
 
 struct spidev_data {
 	dev_t			devt;
diff --git a/tools/spi/Makefile b/tools/spi/Makefile
index 2249a1546cc1..ada881afb489 100644
--- a/tools/spi/Makefile
+++ b/tools/spi/Makefile
@@ -52,7 +52,9 @@ $(OUTPUT)spidev_fdx: $(SPIDEV_FDX_IN)
 clean:
 	rm -f $(ALL_PROGRAMS)
 	rm -rf $(OUTPUT)include/
-	find $(if $(OUTPUT),$(OUTPUT),.) -name '*.o' -delete -o -name '\.*.d' -delete
+	find $(if $(OUTPUT),$(OUTPUT),.) -name '*.o' -delete
+	find $(if $(OUTPUT),$(OUTPUT),.) -name '\.*.o.d' -delete
+	find $(if $(OUTPUT),$(OUTPUT),.) -name '\.*.o.cmd' -delete
 
 install: $(ALL_PROGRAMS)
 	install -d -m 755 $(DESTDIR)$(bindir);		\
diff --git a/tools/spi/spidev_test.c b/tools/spi/spidev_test.c
index 27967dd90f8f..eec23fa693bd 100644
--- a/tools/spi/spidev_test.c
+++ b/tools/spi/spidev_test.c
@@ -128,18 +128,22 @@ static void transfer(int fd, uint8_t const *tx, uint8_t const *rx, size_t len)
 		.bits_per_word = bits,
 	};
 
-	if (mode & SPI_TX_QUAD)
+	if (mode & SPI_TX_OCTAL)
+		tr.tx_nbits = 8;
+	else if (mode & SPI_TX_QUAD)
 		tr.tx_nbits = 4;
 	else if (mode & SPI_TX_DUAL)
 		tr.tx_nbits = 2;
-	if (mode & SPI_RX_QUAD)
+	if (mode & SPI_RX_OCTAL)
+		tr.rx_nbits = 8;
+	else if (mode & SPI_RX_QUAD)
 		tr.rx_nbits = 4;
 	else if (mode & SPI_RX_DUAL)
 		tr.rx_nbits = 2;
 	if (!(mode & SPI_LOOP)) {
-		if (mode & (SPI_TX_QUAD | SPI_TX_DUAL))
+		if (mode & (SPI_TX_OCTAL | SPI_TX_QUAD | SPI_TX_DUAL))
 			tr.rx_buf = 0;
-		else if (mode & (SPI_RX_QUAD | SPI_RX_DUAL))
+		else if (mode & (SPI_RX_OCTAL | SPI_RX_QUAD | SPI_RX_DUAL))
 			tr.tx_buf = 0;
 	}
 
@@ -187,6 +191,7 @@ static void print_usage(const char *prog)
 	     "  -R --ready    slave pulls low to pause\n"
 	     "  -2 --dual     dual transfer\n"
 	     "  -4 --quad     quad transfer\n"
+	     "  -8 --octal    octal transfer\n"
 	     "  -S --size     transfer size\n"
 	     "  -I --iter     iterations\n");
 	exit(1);
@@ -213,13 +218,14 @@ static void parse_opts(int argc, char *argv[])
 			{ "dual",    0, 0, '2' },
 			{ "verbose", 0, 0, 'v' },
 			{ "quad",    0, 0, '4' },
+			{ "octal",   0, 0, '8' },
 			{ "size",    1, 0, 'S' },
 			{ "iter",    1, 0, 'I' },
 			{ NULL, 0, 0, 0 },
 		};
 		int c;
 
-		c = getopt_long(argc, argv, "D:s:d:b:i:o:lHOLC3NR24p:vS:I:",
+		c = getopt_long(argc, argv, "D:s:d:b:i:o:lHOLC3NR248p:vS:I:",
 				lopts, NULL);
 
 		if (c == -1)
@@ -280,6 +286,9 @@ static void parse_opts(int argc, char *argv[])
 		case '4':
 			mode |= SPI_TX_QUAD;
 			break;
+		case '8':
+			mode |= SPI_TX_OCTAL;
+			break;
 		case 'S':
 			transfer_size = atoi(optarg);
 			break;
@@ -295,6 +304,8 @@ static void parse_opts(int argc, char *argv[])
 			mode |= SPI_RX_DUAL;
 		if (mode & SPI_TX_QUAD)
 			mode |= SPI_RX_QUAD;
+		if (mode & SPI_TX_OCTAL)
+			mode |= SPI_RX_OCTAL;
 	}
 }