spi: Updates for v5.16

This is quite a quiet release for SPI, there's been a bit of cleanup to the core from Uwe but nothing functionality wise. We have added several new drivers, Cadence XSPI, Ingenic JZ47xx, Qualcomm SC7280 and SC7180 and Xilinx Versal OSPI. There's a trivial conflict in the Tegra driver that's been causing issues. -----BEGIN PGP SIGNATURE----- iQEzBAABCgAdFiEEreZoqmdXGLWf4p/qJNaLcl1Uh9AFAmGABLQACgkQJNaLcl1U h9Bz2gf9FdjnFXXUQSDgz5tXbrcROCHAYyY9aI9xLFPXaup2ZacKFB5iQcCFaLAR saOoiezNru+Y5MyyEhMcbRhyNeVNwTVven7r2SG6S3ldt3q6RaiDiBr5rUCaTNTj AaatHLrpCfj4d/0Rgzh366BjbAthZQ9+f/c51pEfkoHyY8Kd0XJmi0pCuHnV8reT vzJsGFctcc6Zkzp3M2bg0wG9T+OEJjeMMd/OeYHUFrfhvEsUm0ljRCy1/WRTQ0Fq qlJO/m6YLo4a3D4bl/CH7WwsWrZ/rWEaL2/esyKH799E7MXOOY9D9YdRVU6m+hZS pR1RjqqsXszIMkkviqbPtCA0tF7+4g== =N0Jw -----END PGP SIGNATURE----- Merge tag 'spi-v5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi Pull spi updates from Mark Brown: "This is quite a quiet release for SPI, there's been a bit of cleanup to the core from Uwe but nothing functionality wise. We have added several new drivers, Cadence XSPI, Ingenic JZ47xx, Qualcomm SC7280 and SC7180 and Xilinx Versal OSPI" * tag 'spi-v5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (41 commits) spi: Convert NXP flexspi to json schema spi: spi-geni-qcom: Add support for GPI dma spi: fsi: Fix contention in the FSI2SPI engine spi: spi-rpc-if: Check return value of rpcif_sw_init() spi: tegra210-quad: Put device into suspend on driver removal spi: tegra20-slink: Put device into suspend on driver removal spi: bcm-qspi: Fix missing clk_disable_unprepare() on error in bcm_qspi_probe() spi: at91-usart: replacing legacy gpio interface for gpiod spi: replace snprintf in show functions with sysfs_emit spi: cadence: Add of_node_put() before return spi: orion: Add of_node_put() before goto spi: cadence-quadspi: fix dma_unmap_single() call spi: tegra20: fix build with CONFIG_PM_SLEEP=n spi: bcm-qspi: add support for 3-wire mode for half duplex transfer spi: bcm-qspi: Add mspi spcr3 32/64-bits xfer mode spi: Make several public functions private to spi.c spi: Reorder functions to simplify the next commit spi: Remove unused function spi_busnum_to_master() spi: Move comment about chipselect check to the right place spi: fsi: Print status on error ...
2021-11-01 19:09:04 -07:00 · 2021-11-01 19:09:04 -07:00 · 2019295c9e
--- a/Documentation/devicetree/bindings/spi/cdns,qspi-nor.yaml
+++ b/Documentation/devicetree/bindings/spi/cdns,qspi-nor.yaml
@ -11,6 +11,14 @@ maintainers:

 allOf:
  - $ref: spi-controller.yaml#
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: xlnx,versal-ospi-1.0
+    then:
+      required:
+        - power-domains

 properties:
  compatible:
@ -20,6 +28,7 @@ properties:
              - ti,k2g-qspi
              - ti,am654-ospi
              - intel,lgm-qspi
+              - xlnx,versal-ospi-1.0
          - const: cdns,qspi-nor
      - const: cdns,qspi-nor

@ -65,6 +74,9 @@ properties:
      data rather than the QSPI clock. Make sure that QSPI return clock
      is populated on the board before using this property.

+  power-domains:
+    maxItems: 1
+
  resets:
    maxItems: 2

--- a/Documentation/devicetree/bindings/spi/cdns,xspi.yaml
+++ b/Documentation/devicetree/bindings/spi/cdns,xspi.yaml
@ -0,0 +1,77 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+# Copyright 2020-21 Cadence
+%YAML 1.2
+---
+$id: "http://devicetree.org/schemas/spi/cdns,xspi.yaml#"
+$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+
+title: Cadence XSPI Controller
+
+maintainers:
+  - Parshuram Thombare <pthombar@cadence.com>
+
+description: |
+  The XSPI controller allows SPI protocol communication in
+  single, dual, quad or octal wire transmission modes for
+  read/write access to slaves such as SPI-NOR flash.
+
+allOf:
+  - $ref: "spi-controller.yaml#"
+
+properties:
+  compatible:
+    const: cdns,xspi-nor
+
+  reg:
+    items:
+      - description: address and length of the controller register set
+      - description: address and length of the Slave DMA data port
+      - description: address and length of the auxiliary registers
+
+  reg-names:
+    items:
+      - const: io
+      - const: sdma
+      - const: aux
+
+  interrupts:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+  - interrupts
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+    bus {
+        #address-cells = <2>;
+        #size-cells = <2>;
+
+        xspi: spi@a0010000 {
+            #address-cells = <1>;
+            #size-cells = <0>;
+            compatible = "cdns,xspi-nor";
+            reg = <0x0 0xa0010000 0x0 0x1040>,
+                  <0x0 0xb0000000 0x0 0x1000>,
+                  <0x0 0xa0020000 0x0 0x100>;
+            reg-names = "io", "sdma", "aux";
+            interrupts = <0 90 IRQ_TYPE_LEVEL_HIGH>;
+            interrupt-parent = <&gic>;
+
+            flash@0 {
+                compatible = "jedec,spi-nor";
+                spi-max-frequency = <75000000>;
+                reg = <0>;
+            };
+
+            flash@1 {
+                compatible = "jedec,spi-nor";
+                spi-max-frequency = <75000000>;
+                reg = <1>;
+            };
+        };
+    };
--- a/Documentation/devicetree/bindings/spi/ingenic,spi.yaml
+++ b/Documentation/devicetree/bindings/spi/ingenic,spi.yaml
@ -0,0 +1,72 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/spi/ingenic,spi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Ingenic SoCs SPI controller devicetree bindings
+
+maintainers:
+  - Artur Rojek <contact@artur-rojek.eu>
+  - Paul Cercueil <paul@crapouillou.net>
+
+allOf:
+  - $ref: /schemas/spi/spi-controller.yaml#
+
+properties:
+  compatible:
+    oneOf:
+      - enum:
+          - ingenic,jz4750-spi
+          - ingenic,jz4780-spi
+      - items:
+          - enum:
+              - ingenic,jz4760-spi
+              - ingenic,jz4770-spi
+          - const: ingenic,jz4750-spi
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    maxItems: 1
+
+  dmas:
+    maxItems: 2
+    minItems: 2
+
+  dma-names:
+    items:
+      - const: rx
+      - const: tx
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - clocks
+  - dmas
+  - dma-names
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/jz4770-cgu.h>
+    spi@10043000 {
+      compatible = "ingenic,jz4770-spi", "ingenic,jz4750-spi";
+      reg = <0x10043000 0x1c>;
+      #address-cells = <1>;
+      #size-cells = <0>;
+
+      interrupt-parent = <&intc>;
+      interrupts = <8>;
+
+      clocks = <&cgu JZ4770_CLK_SSI0>;
+
+      dmas = <&dmac1 23 0xffffffff>, <&dmac1 22 0xffffffff>;
+      dma-names = "rx", "tx";
+    };
--- a/Documentation/devicetree/bindings/spi/qcom,spi-qcom-qspi.yaml
+++ b/Documentation/devicetree/bindings/spi/qcom,spi-qcom-qspi.yaml
@ -21,7 +21,11 @@ allOf:
 properties:
  compatible:
    items:
-      - const: qcom,sdm845-qspi
+      - enum:
+          - qcom,sc7180-qspi
+          - qcom,sc7280-qspi
+          - qcom,sdm845-qspi
+
      - const: qcom,qspi-v1

  reg:
--- a/Documentation/devicetree/bindings/spi/spi-nxp-fspi.txt
+++ b/Documentation/devicetree/bindings/spi/spi-nxp-fspi.txt
@ -1,44 +0,0 @@
-* NXP Flex Serial Peripheral Interface (FSPI)
-
-Required properties:
-  - compatible : Should be "nxp,lx2160a-fspi"
-			    "nxp,imx8qxp-fspi"
-			    "nxp,imx8mm-fspi"
-			    "nxp,imx8mp-fspi"
-			    "nxp,imx8dxl-fspi"
-
-  - reg :        First contains the register location and length,
-                 Second contains the memory mapping address and length
-  - reg-names :  Should contain the resource reg names:
-	         - fspi_base: configuration register address space
-                 - fspi_mmap: memory mapped address space
-  - interrupts : Should contain the interrupt for the device
-
-Required SPI slave node properties:
-  - reg :        There are two buses (A and B) with two chip selects each.
-                 This encodes to which bus and CS the flash is connected:
-                 - <0>: Bus A, CS 0
-                 - <1>: Bus A, CS 1
-                 - <2>: Bus B, CS 0
-                 - <3>: Bus B, CS 1
-
-Example showing the usage of two SPI NOR slave devices on bus A:
-
-fspi0: spi@20c0000 {
-	compatible = "nxp,lx2160a-fspi";
-	reg = <0x0 0x20c0000 0x0 0x10000>, <0x0 0x20000000 0x0 0x10000000>;
-	reg-names = "fspi_base", "fspi_mmap";
-	interrupts = <0 25 0x4>; /* Level high type */
-	clocks = <&clockgen 4 3>, <&clockgen 4 3>;
-	clock-names = "fspi_en", "fspi";
-
-	mt35xu512aba0: flash@0 {
-		reg = <0>;
-		....
-	};
-
-	mt35xu512aba1: flash@1 {
-		reg = <1>;
-		....
-	};
-};
--- a/Documentation/devicetree/bindings/spi/spi-nxp-fspi.yaml
+++ b/Documentation/devicetree/bindings/spi/spi-nxp-fspi.yaml
@ -0,0 +1,86 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/spi/spi-nxp-fspi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: NXP Flex Serial Peripheral Interface (FSPI)
+
+maintainers:
+  - Kuldeep Singh <kuldeep.singh@nxp.com>
+
+allOf:
+  - $ref: "spi-controller.yaml#"
+
+properties:
+  compatible:
+    enum:
+      - nxp,imx8dxl-fspi
+      - nxp,imx8mm-fspi
+      - nxp,imx8mp-fspi
+      - nxp,imx8qxp-fspi
+      - nxp,lx2160a-fspi
+
+  reg:
+    items:
+      - description: registers address space
+      - description: memory mapped address space
+
+  reg-names:
+    items:
+      - const: fspi_base
+      - const: fspi_mmap
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    items:
+      - description: SPI bus clock
+      - description: SPI serial clock
+
+  clock-names:
+    items:
+      - const: fspi_en
+      - const: fspi
+
+required:
+  - compatible
+  - reg
+  - reg-names
+  - interrupts
+  - clocks
+  - clock-names
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/clock/fsl,qoriq-clockgen.h>
+
+    soc {
+        #address-cells = <2>;
+        #size-cells = <2>;
+
+        spi@20c0000 {
+            compatible = "nxp,lx2160a-fspi";
+            reg = <0x0 0x20c0000 0x0 0x100000>,
+                  <0x0 0x20000000 0x0 0x10000000>;
+            reg-names = "fspi_base", "fspi_mmap";
+            interrupts = <GIC_SPI 25 IRQ_TYPE_LEVEL_HIGH>;
+            clocks = <&clockgen QORIQ_CLK_PLATFORM_PLL QORIQ_CLK_PLL_DIV(4)>,
+                     <&clockgen QORIQ_CLK_PLATFORM_PLL QORIQ_CLK_PLL_DIV(4)>;
+            clock-names = "fspi_en", "fspi";
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            flash@0 {
+                compatible = "jedec,spi-nor";
+                spi-max-frequency = <50000000>;
+                reg = <0>;
+                spi-rx-bus-width = <8>;
+                spi-tx-bus-width = <8>;
+            };
+        };
+    };
--- a/Documentation/spi/spi-summary.rst
+++ b/Documentation/spi/spi-summary.rst
@ -336,14 +336,6 @@ certainly includes SPI devices hooked up through the card connectors!
 Non-static Configurations
 ^^^^^^^^^^^^^^^^^^^^^^^^^

-Developer boards often play by different rules than product boards, and one
-example is the potential need to hotplug SPI devices and/or controllers.
-
-For those cases you might need to use spi_busnum_to_master() to look
-up the spi bus master, and will likely need spi_new_device() to provide the
-board info based on the board that was hotplugged.  Of course, you'd later
-call at least spi_unregister_device() when that board is removed.
-
 When Linux includes support for MMC/SD/SDIO/DataFlash cards through SPI, those
 configurations will also be dynamic.  Fortunately, such devices all support
 basic device identification probes, so they should hotplug normally.
--- a/2
+++ b/2
@ -13488,7 +13488,7 @@ M:	Ashish Kumar <ashish.kumar@nxp.com>
 R:	Yogesh Gaur <yogeshgaur.83@gmail.com>
 L:	linux-spi@vger.kernel.org
 S:	Maintained
-F:	Documentation/devicetree/bindings/spi/spi-nxp-fspi.txt
+F:	Documentation/devicetree/bindings/spi/spi-nxp-fspi.yaml
 F:	drivers/spi/spi-nxp-fspi.c

 NXP FXAS21002C DRIVER
--- a/arch/mips/boot/dts/ingenic/ci20.dts
+++ b/arch/mips/boot/dts/ingenic/ci20.dts
@ -113,9 +113,12 @@
 	 * Use the 32.768 kHz oscillator as the parent of the RTC for a higher
 	 * precision.
 	 */
-	assigned-clocks = <&cgu JZ4780_CLK_OTGPHY>, <&cgu JZ4780_CLK_RTC>;
-	assigned-clock-parents = <0>, <&cgu JZ4780_CLK_RTCLK>;
-	assigned-clock-rates = <48000000>;
+	assigned-clocks = <&cgu JZ4780_CLK_OTGPHY>, <&cgu JZ4780_CLK_RTC>,
+			  <&cgu JZ4780_CLK_SSIPLL>, <&cgu JZ4780_CLK_SSI>;
+	assigned-clock-parents = <0>, <&cgu JZ4780_CLK_RTCLK>,
+				 <&cgu JZ4780_CLK_MPLL>,
+				 <&cgu JZ4780_CLK_SSIPLL>;
+	assigned-clock-rates = <48000000>, <0>, <54000000>;
 };

 &tcu {
--- a/arch/mips/boot/dts/ingenic/jz4780.dtsi
+++ b/arch/mips/boot/dts/ingenic/jz4780.dtsi
@ -255,22 +255,23 @@
 		};
 	};

-	spi_gpio {
-		compatible = "spi-gpio";
+	spi0: spi@10043000 {
+		compatible = "ingenic,jz4780-spi";
+		reg = <0x10043000 0x1c>;
 		#address-cells = <1>;
 		#size-cells = <0>;
-		num-chipselects = <2>;

-		gpio-miso = <&gpe 14 0>;
-		gpio-sck = <&gpe 15 0>;
-		gpio-mosi = <&gpe 17 0>;
-		cs-gpios = <&gpe 16 0>, <&gpe 18 0>;
+		interrupt-parent = <&intc>;
+		interrupts = <8>;

-		spidev@0 {
-			compatible = "spidev";
-			reg = <0>;
-			spi-max-frequency = <1000000>;
-		};
+		clocks = <&cgu JZ4780_CLK_SSI0>;
+		clock-names = "spi";
+
+		dmas = <&dma JZ4780_DMA_SSI0_RX 0xffffffff>,
+		       <&dma JZ4780_DMA_SSI0_TX 0xffffffff>;
+		dma-names = "rx", "tx";
+
+		status = "disabled";
 	};

 	uart0: serial@10030000 {
@ -338,6 +339,25 @@
 		status = "disabled";
 	};

+	spi1: spi@10044000 {
+		compatible = "ingenic,jz4780-spi";
+		reg = <0x10044000 0x1c>;
+		#address-cells = <1>;
+		#size-sells = <0>;
+
+		interrupt-parent = <&intc>;
+		interrupts = <7>;
+
+		clocks = <&cgu JZ4780_CLK_SSI1>;
+		clock-names = "spi";
+
+		dmas = <&dma JZ4780_DMA_SSI1_RX 0xffffffff>,
+		       <&dma JZ4780_DMA_SSI1_TX 0xffffffff>;
+		dma-names = "rx", "tx";
+
+		status = "disabled";
+	};
+
 	i2c0: i2c@10050000 {
 		compatible = "ingenic,jz4780-i2c", "ingenic,jz4770-i2c";
 		#address-cells = <1>;
--- a/drivers/firmware/xilinx/zynqmp.c
+++ b/drivers/firmware/xilinx/zynqmp.c
@ -647,6 +647,23 @@ int zynqmp_pm_sd_dll_reset(u32 node_id, u32 type)
 }
 EXPORT_SYMBOL_GPL(zynqmp_pm_sd_dll_reset);

+/**
+ * zynqmp_pm_ospi_mux_select() - OSPI Mux selection
+ *
+ * @dev_id:	Device Id of the OSPI device.
+ * @select:	OSPI Mux select value.
+ *
+ * This function select the OSPI Mux.
+ *
+ * Return:	Returns status, either success or error+reason
+ */
+int zynqmp_pm_ospi_mux_select(u32 dev_id, u32 select)
+{
+	return zynqmp_pm_invoke_fn(PM_IOCTL, dev_id, IOCTL_OSPI_MUX_SELECT,
+				   select, 0, NULL);
+}
+EXPORT_SYMBOL_GPL(zynqmp_pm_ospi_mux_select);
+
 /**
 * zynqmp_pm_write_ggs() - PM API for writing global general storage (ggs)
 * @index:	GGS register index
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@ -228,6 +228,18 @@ config SPI_CADENCE_QUADSPI
 	  device with a Cadence QSPI controller and want to access the
 	  Flash as an MTD device.

+config SPI_CADENCE_XSPI
+	tristate "Cadence XSPI controller"
+	depends on (OF || COMPILE_TEST) && HAS_IOMEM
+	depends on SPI_MEM
+	help
+	  Enable support for the Cadence XSPI Flash controller.
+
+	  Cadence XSPI is a specialized controller for connecting an SPI
+	  Flash over upto 8bit wide bus. Enable this option if you have a
+	  device with a Cadence XSPI controller and want to access the
+	  Flash as an MTD device.
+
 config SPI_CLPS711X
 	tristate "CLPS711X host SPI controller"
 	depends on ARCH_CLPS711X || COMPILE_TEST
@ -406,6 +418,15 @@ config SPI_IMX
 	help
 	  This enables support for the Freescale i.MX SPI controllers.

+config SPI_INGENIC
+	tristate "Ingenic JZ47xx SoCs SPI controller"
+	depends on MACH_INGENIC || COMPILE_TEST
+	help
+	  This enables support for the Ingenic JZ47xx SoCs SPI controller.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called spi-ingenic.
+
 config SPI_JCORE
 	tristate "J-Core SPI Master"
 	depends on OF && (SUPERH || COMPILE_TEST)
@ -738,10 +759,11 @@ config SPI_S3C24XX_FIQ
 	  TX and RX data paths.

 config SPI_S3C64XX
-	tristate "Samsung S3C64XX series type SPI"
+	tristate "Samsung S3C64XX/Exynos SoC series type SPI"
 	depends on (PLAT_SAMSUNG || ARCH_S5PV210 || ARCH_EXYNOS || COMPILE_TEST)
 	help
-	  SPI driver for Samsung S3C64XX and newer SoCs.
+	  SPI driver for Samsung S3C64XX, S5Pv210 and Exynos SoCs.
+	  Choose Y/M here only if you build for such Samsung SoC.

 config SPI_SC18IS602
 	tristate "NXP SC18IS602/602B/603 I2C to SPI bridge"
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@ -34,6 +34,7 @@ obj-$(CONFIG_SPI_BITBANG)		+= spi-bitbang.o
 obj-$(CONFIG_SPI_BUTTERFLY)		+= spi-butterfly.o
 obj-$(CONFIG_SPI_CADENCE)		+= spi-cadence.o
 obj-$(CONFIG_SPI_CADENCE_QUADSPI)	+= spi-cadence-quadspi.o
+obj-$(CONFIG_SPI_CADENCE_XSPI)		+= spi-cadence-xspi.o
 obj-$(CONFIG_SPI_CLPS711X)		+= spi-clps711x.o
 obj-$(CONFIG_SPI_COLDFIRE_QSPI)		+= spi-coldfire-qspi.o
 obj-$(CONFIG_SPI_DAVINCI)		+= spi-davinci.o
@ -59,6 +60,7 @@ obj-$(CONFIG_SPI_HISI_KUNPENG)		+= spi-hisi-kunpeng.o
 obj-$(CONFIG_SPI_HISI_SFC_V3XX)		+= spi-hisi-sfc-v3xx.o
 obj-$(CONFIG_SPI_IMG_SPFI)		+= spi-img-spfi.o
 obj-$(CONFIG_SPI_IMX)			+= spi-imx.o
+obj-$(CONFIG_SPI_INGENIC)		+= spi-ingenic.o
 obj-$(CONFIG_SPI_LANTIQ_SSC)		+= spi-lantiq-ssc.o
 obj-$(CONFIG_SPI_JCORE)			+= spi-jcore.o
 obj-$(CONFIG_SPI_LM70_LLP)		+= spi-lm70llp.o
--- a/drivers/spi/atmel-quadspi.c
+++ b/drivers/spi/atmel-quadspi.c
@ -310,7 +310,7 @@ static int atmel_qspi_set_cfg(struct atmel_qspi *aq,
 		return mode;
 	ifr |= atmel_qspi_modes[mode].config;

-	if (op->dummy.buswidth && op->dummy.nbytes)
+	if (op->dummy.nbytes)
 		dummy_cycles = op->dummy.nbytes * 8 / op->dummy.buswidth;

 	/*
--- a/drivers/spi/spi-amd.c
+++ b/drivers/spi/spi-amd.c
@ -38,126 +38,102 @@ 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)
+static inline u8 amd_spi_readreg8(struct amd_spi *amd_spi, 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)
+static inline void amd_spi_writereg8(struct amd_spi *amd_spi, 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)
+static void amd_spi_setclear_reg8(struct amd_spi *amd_spi, int idx, u8 set, u8 clear)
 {
-	u8 tmp = amd_spi_readreg8(master, idx);
+	u8 tmp = amd_spi_readreg8(amd_spi, idx);

 	tmp = (tmp & ~clear) | set;
-	amd_spi_writereg8(master, idx, tmp);
+	amd_spi_writereg8(amd_spi, idx, tmp);
 }

-static inline u32 amd_spi_readreg32(struct spi_master *master, int idx)
+static inline u32 amd_spi_readreg32(struct amd_spi *amd_spi, 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)
+static inline void amd_spi_writereg32(struct amd_spi *amd_spi, 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)
+static inline void amd_spi_setclear_reg32(struct amd_spi *amd_spi, int idx, u32 set, u32 clear)
 {
-	u32 tmp = amd_spi_readreg32(master, idx);
+	u32 tmp = amd_spi_readreg32(amd_spi, idx);

 	tmp = (tmp & ~clear) | set;
-	amd_spi_writereg32(master, idx, tmp);
+	amd_spi_writereg32(amd_spi, idx, tmp);
 }

-static void amd_spi_select_chip(struct spi_master *master)
+static void amd_spi_select_chip(struct amd_spi *amd_spi, u8 cs)
 {
-	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);
+	amd_spi_setclear_reg8(amd_spi, AMD_SPI_ALT_CS_REG, cs, AMD_SPI_ALT_CS_MASK);
 }

-static void amd_spi_clear_fifo_ptr(struct spi_master *master)
+static void amd_spi_clear_fifo_ptr(struct amd_spi *amd_spi)
 {
-	amd_spi_setclear_reg32(master, AMD_SPI_CTRL0_REG, AMD_SPI_FIFO_CLEAR,
-			       AMD_SPI_FIFO_CLEAR);
+	amd_spi_setclear_reg32(amd_spi, 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)
+static void amd_spi_set_opcode(struct amd_spi *amd_spi, u8 cmd_opcode)
 {
-	amd_spi_setclear_reg32(master, AMD_SPI_CTRL0_REG, cmd_opcode,
-			       AMD_SPI_OPCODE_MASK);
+	amd_spi_setclear_reg32(amd_spi, 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)
+static inline void amd_spi_set_rx_count(struct amd_spi *amd_spi, u8 rx_count)
 {
-	amd_spi_setclear_reg8(master, AMD_SPI_RX_COUNT_REG, rx_count, 0xff);
+	amd_spi_setclear_reg8(amd_spi, AMD_SPI_RX_COUNT_REG, rx_count, 0xff);
 }

-static inline void amd_spi_set_tx_count(struct spi_master *master,
-					u8 tx_count)
+static inline void amd_spi_set_tx_count(struct amd_spi *amd_spi, u8 tx_count)
 {
-	amd_spi_setclear_reg8(master, AMD_SPI_TX_COUNT_REG, tx_count, 0xff);
+	amd_spi_setclear_reg8(amd_spi, AMD_SPI_TX_COUNT_REG, tx_count, 0xff);
 }

-static inline int amd_spi_busy_wait(struct amd_spi *amd_spi)
+static 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) {
+	while (amd_spi_readreg32(amd_spi, AMD_SPI_CTRL0_REG) & AMD_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)
+static int amd_spi_execute_opcode(struct amd_spi *amd_spi)
 {
-	struct amd_spi *amd_spi = spi_master_get_devdata(master);
+	int ret;
+
+	ret = amd_spi_busy_wait(amd_spi);
+	if (ret)
+		return ret;

 	/* 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_setclear_reg32(amd_spi, AMD_SPI_CTRL0_REG, AMD_SPI_EXEC_CMD, AMD_SPI_EXEC_CMD);

-	amd_spi_busy_wait(amd_spi);
+	return 0;
 }

 static int amd_spi_master_setup(struct spi_device *spi)
 {
-	struct spi_master *master = spi->master;
+	struct amd_spi *amd_spi = spi_master_get_devdata(spi->master);

-	amd_spi_clear_fifo_ptr(master);
+	amd_spi_clear_fifo_ptr(amd_spi);

 	return 0;
 }
@ -185,19 +161,18 @@ static inline int amd_spi_fifo_xfer(struct amd_spi *amd_spi,
 			tx_len = xfer->len - 1;
 			cmd_opcode = *(u8 *)xfer->tx_buf;
 			buf++;
-			amd_spi_set_opcode(master, cmd_opcode);
+			amd_spi_set_opcode(amd_spi, cmd_opcode);

 			/* Write data into the FIFO. */
 			for (i = 0; i < tx_len; i++) {
-				iowrite8(buf[i],
-					 ((u8 __iomem *)amd_spi->io_remap_addr +
+				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);
+			amd_spi_set_tx_count(amd_spi, tx_len);
+			amd_spi_clear_fifo_ptr(amd_spi);
 			/* Execute command */
-			amd_spi_execute_opcode(master);
+			amd_spi_execute_opcode(amd_spi);
 		}
 		if (m_cmd & AMD_SPI_XFER_RX) {
 			/*
@ -206,15 +181,14 @@ static inline int amd_spi_fifo_xfer(struct amd_spi *amd_spi,
 			 */
 			rx_len = xfer->len;
 			buf = (u8 *)xfer->rx_buf;
-			amd_spi_set_rx_count(master, rx_len);
-			amd_spi_clear_fifo_ptr(master);
+			amd_spi_set_rx_count(amd_spi, rx_len);
+			amd_spi_clear_fifo_ptr(amd_spi);
 			/* Execute command */
-			amd_spi_execute_opcode(master);
+			amd_spi_execute_opcode(amd_spi);
+			amd_spi_busy_wait(amd_spi);
 			/* 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);
+				buf[i] = amd_spi_readreg8(amd_spi, AMD_SPI_FIFO_BASE + tx_len + i);
 		}
 	}

@ -233,8 +207,7 @@ static int amd_spi_master_transfer(struct spi_master *master,
 	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);
+	amd_spi_select_chip(amd_spi, spi->chip_select);

 	/*
 	 * Extract spi_transfers from the spi message and
--- a/drivers/spi/spi-at91-usart.c
+++ b/drivers/spi/spi-at91-usart.c
@ -14,7 +14,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/of_platform.h>
-#include <linux/of_gpio.h>
+#include <linux/gpio/consumer.h>
 #include <linux/pinctrl/consumer.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
@ -482,29 +482,12 @@ static void at91_usart_spi_init(struct at91_usart_spi *aus)

 static int at91_usart_gpio_setup(struct platform_device *pdev)
 {
-	struct device_node *np = pdev->dev.parent->of_node;
-	int i;
-	int ret;
-	int nb;
+	struct gpio_descs *cs_gpios;

-	if (!np)
-		return -EINVAL;
+	cs_gpios = devm_gpiod_get_array_optional(&pdev->dev, "cs", GPIOD_OUT_LOW);

-	nb = of_gpio_named_count(np, "cs-gpios");
-	for (i = 0; i < nb; i++) {
-		int cs_gpio = of_get_named_gpio(np, "cs-gpios", i);
-
-		if (cs_gpio < 0)
-			return cs_gpio;
-
-		if (gpio_is_valid(cs_gpio)) {
-			ret = devm_gpio_request_one(&pdev->dev, cs_gpio,
-						    GPIOF_DIR_OUT,
-						    dev_name(&pdev->dev));
-			if (ret)
-				return ret;
-		}
-	}
+	if (IS_ERR(cs_gpios))
+		return PTR_ERR(cs_gpios);

 	return 0;
 }
--- a/drivers/spi/spi-bcm-qspi.c
+++ b/drivers/spi/spi-bcm-qspi.c
@ -83,6 +83,9 @@
 /* MSPI register offsets */
 #define MSPI_SPCR0_LSB				0x000
 #define MSPI_SPCR0_MSB				0x004
+#define MSPI_SPCR0_MSB_CPHA			BIT(0)
+#define MSPI_SPCR0_MSB_CPOL			BIT(1)
+#define MSPI_SPCR0_MSB_BITS_SHIFT		0x2
 #define MSPI_SPCR1_LSB				0x008
 #define MSPI_SPCR1_MSB				0x00c
 #define MSPI_NEWQP				0x010
@ -100,8 +103,10 @@
 #define MSPI_MASTER_BIT			BIT(7)

 #define MSPI_NUM_CDRAM				16
+#define MSPI_CDRAM_OUTP				BIT(8)
 #define MSPI_CDRAM_CONT_BIT			BIT(7)
 #define MSPI_CDRAM_BITSE_BIT			BIT(6)
+#define MSPI_CDRAM_DT_BIT			BIT(5)
 #define MSPI_CDRAM_PCS				0xf

 #define MSPI_SPCR2_SPE				BIT(6)
@ -114,6 +119,14 @@
 						 ~(BIT(10) | BIT(11)))
 #define MSPI_SPCR3_SYSCLKSEL_108		(MSPI_SPCR3_SYSCLKSEL_MASK & \
 						 BIT(11))
+#define MSPI_SPCR3_TXRXDAM_MASK			GENMASK(4, 2)
+#define MSPI_SPCR3_DAM_8BYTE			0
+#define MSPI_SPCR3_DAM_16BYTE			(BIT(2) | BIT(4))
+#define MSPI_SPCR3_DAM_32BYTE			(BIT(3) | BIT(5))
+#define MSPI_SPCR3_HALFDUPLEX			BIT(6)
+#define MSPI_SPCR3_HDOUTTYPE			BIT(7)
+#define MSPI_SPCR3_DATA_REG_SZ			BIT(8)
+#define MSPI_SPCR3_CPHARX			BIT(9)

 #define MSPI_MSPI_STATUS_SPIF			BIT(0)

@ -153,6 +166,14 @@
 #define TRANS_STATUS_BREAK_DESELECT (TRANS_STATUS_BREAK_EOM |		\
 				     TRANS_STATUS_BREAK_CS_CHANGE)

+/*
+ * Used for writing and reading data in the right order
+ * to TXRAM and RXRAM when used as 32-bit registers respectively
+ */
+#define swap4bytes(__val) \
+	((((__val) >> 24) & 0x000000FF) | (((__val) >>  8) & 0x0000FF00) | \
+	 (((__val) <<  8) & 0x00FF0000) | (((__val) << 24) & 0xFF000000))
+
 struct bcm_qspi_parms {
 	u32 speed_hz;
 	u8 mode;
@ -261,7 +282,7 @@ static inline bool bcm_qspi_has_sysclk_108(struct bcm_qspi *qspi)
 static inline int bcm_qspi_spbr_min(struct bcm_qspi *qspi)
 {
 	if (bcm_qspi_has_fastbr(qspi))
-		return 1;
+		return (bcm_qspi_has_sysclk_108(qspi) ? 4 : 1);
 	else
 		return 8;
 }
@ -395,6 +416,7 @@ static int bcm_qspi_bspi_set_flex_mode(struct bcm_qspi *qspi,
 	if (addrlen == BSPI_ADDRLEN_4BYTES)
 		bpp = BSPI_BPP_ADDR_SELECT_MASK;

+	if (op->dummy.nbytes)
 		bpp |= (op->dummy.nbytes * 8) / op->dummy.buswidth;

 	switch (width) {
@ -570,23 +592,23 @@ static void bcm_qspi_hw_set_parms(struct bcm_qspi *qspi,
 {
 	u32 spcr, spbr = 0;

-	if (xp->speed_hz)
-		spbr = qspi->base_clk / (2 * xp->speed_hz);
-
-	spcr = clamp_val(spbr, bcm_qspi_spbr_min(qspi), QSPI_SPBR_MAX);
-	bcm_qspi_write(qspi, MSPI, MSPI_SPCR0_LSB, spcr);
-
 	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;
+	/*
+	 * Bits per transfer.  BITS determines the number of data bits
+	 * transferred if the command control bit (BITSE of a
+	 * CDRAM Register) is equal to 1.
+	 * If CDRAM BITSE is equal to 0, 8 data bits are transferred
+	 * regardless
+	 */
+	if (xp->bits_per_word != 16 && xp->bits_per_word != 64)
+		spcr |= xp->bits_per_word << MSPI_SPCR0_MSB_BITS_SHIFT;

+	spcr |= xp->mode & (MSPI_SPCR0_MSB_CPHA | MSPI_SPCR0_MSB_CPOL);
 	bcm_qspi_write(qspi, MSPI, MSPI_SPCR0_MSB, spcr);

 	if (bcm_qspi_has_fastbr(qspi)) {
@ -595,17 +617,44 @@ static void bcm_qspi_hw_set_parms(struct bcm_qspi *qspi,
 		/* enable fastbr */
 		spcr |=	MSPI_SPCR3_FASTBR;

+		if (xp->mode & SPI_3WIRE)
+			spcr |= MSPI_SPCR3_HALFDUPLEX |  MSPI_SPCR3_HDOUTTYPE;
+
 		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);
 		}

+		if (xp->bits_per_word > 16) {
+			/* data_reg_size 1 (64bit) */
+			spcr |=	MSPI_SPCR3_DATA_REG_SZ;
+			/* TxRx RAM data access mode 2 for 32B and set fastdt */
+			spcr |=	MSPI_SPCR3_DAM_32BYTE  | MSPI_SPCR3_FASTDT;
+			/*
+			 *  Set length of delay after transfer
+			 *  DTL from 0(256) to 1
+			 */
+			bcm_qspi_write(qspi, MSPI, MSPI_SPCR1_LSB, 1);
+		} else {
+			/* data_reg_size[8] = 0 */
+			spcr &=	~(MSPI_SPCR3_DATA_REG_SZ);
+
+			/*
+			 * TxRx RAM access mode 8B
+			 * and disable fastdt
+			 */
+			spcr &= ~(MSPI_SPCR3_DAM_32BYTE);
+		}
 		bcm_qspi_write(qspi, MSPI, MSPI_SPCR3, spcr);
 	}

+	if (xp->speed_hz)
+		spbr = qspi->base_clk / (2 * xp->speed_hz);
+
+	spbr = clamp_val(spbr, bcm_qspi_spbr_min(qspi), QSPI_SPBR_MAX);
+	bcm_qspi_write(qspi, MSPI, MSPI_SPCR0_LSB, spbr);
+
 	qspi->last_parms = *xp;
 }

@ -626,7 +675,7 @@ static int bcm_qspi_setup(struct spi_device *spi)
 {
 	struct bcm_qspi_parms *xp;

-	if (spi->bits_per_word > 16)
+	if (spi->bits_per_word > 64)
 		return -EINVAL;

 	xp = spi_get_ctldata(spi);
@ -665,8 +714,12 @@ static int update_qspi_trans_byte_count(struct bcm_qspi *qspi,
 	/* count the last transferred bytes */
 	if (qt->trans->bits_per_word <= 8)
 		qt->byte++;
-	else
+	else if (qt->trans->bits_per_word <= 16)
 		qt->byte += 2;
+	else if (qt->trans->bits_per_word <= 32)
+		qt->byte += 4;
+	else if (qt->trans->bits_per_word <= 64)
+		qt->byte += 8;

 	if (qt->byte >= qt->trans->len) {
 		/* we're at the end of the spi_transfer */
@ -709,6 +762,33 @@ static inline u16 read_rxram_slot_u16(struct bcm_qspi *qspi, int slot)
 		((bcm_qspi_read(qspi, MSPI, msb_offset) & 0xff) << 8);
 }

+static inline u32 read_rxram_slot_u32(struct bcm_qspi *qspi, int slot)
+{
+	u32 reg_offset = MSPI_RXRAM;
+	u32 offset = reg_offset + (slot << 3);
+	u32 val;
+
+	val = bcm_qspi_read(qspi, MSPI, offset);
+	val = swap4bytes(val);
+
+	return val;
+}
+
+static inline u64 read_rxram_slot_u64(struct bcm_qspi *qspi, int slot)
+{
+	u32 reg_offset = MSPI_RXRAM;
+	u32 lsb_offset = reg_offset + (slot << 3) + 0x4;
+	u32 msb_offset = reg_offset + (slot << 3);
+	u32 msb, lsb;
+
+	msb = bcm_qspi_read(qspi, MSPI, msb_offset);
+	msb = swap4bytes(msb);
+	lsb = bcm_qspi_read(qspi, MSPI, lsb_offset);
+	lsb = swap4bytes(lsb);
+
+	return ((u64)msb << 32 | lsb);
+}
+
 static void read_from_hw(struct bcm_qspi *qspi, int slots)
 {
 	struct qspi_trans tp;
@ -732,7 +812,7 @@ static void read_from_hw(struct bcm_qspi *qspi, int slots)
 				buf[tp.byte] = read_rxram_slot_u8(qspi, slot);
 			dev_dbg(&qspi->pdev->dev, "RD %02x\n",
 				buf ? buf[tp.byte] : 0x0);
-		} else {
+		} else if (tp.trans->bits_per_word <= 16) {
 			u16 *buf = tp.trans->rx_buf;

 			if (buf)
@ -740,6 +820,25 @@ static void read_from_hw(struct bcm_qspi *qspi, int slots)
 								      slot);
 			dev_dbg(&qspi->pdev->dev, "RD %04x\n",
 				buf ? buf[tp.byte / 2] : 0x0);
+		} else if (tp.trans->bits_per_word <= 32) {
+			u32 *buf = tp.trans->rx_buf;
+
+			if (buf)
+				buf[tp.byte / 4] = read_rxram_slot_u32(qspi,
+								      slot);
+			dev_dbg(&qspi->pdev->dev, "RD %08x\n",
+				buf ? buf[tp.byte / 4] : 0x0);
+
+		} else if (tp.trans->bits_per_word <= 64) {
+			u64 *buf = tp.trans->rx_buf;
+
+			if (buf)
+				buf[tp.byte / 8] = read_rxram_slot_u64(qspi,
+								      slot);
+			dev_dbg(&qspi->pdev->dev, "RD %llx\n",
+				buf ? buf[tp.byte / 8] : 0x0);
+
+
 		}

 		update_qspi_trans_byte_count(qspi, &tp,
@ -769,6 +868,28 @@ static inline void write_txram_slot_u16(struct bcm_qspi *qspi, int slot,
 	bcm_qspi_write(qspi, MSPI, lsb_offset, (val & 0xff));
 }

+static inline void write_txram_slot_u32(struct bcm_qspi *qspi, int slot,
+					u32 val)
+{
+	u32 reg_offset = MSPI_TXRAM;
+	u32 msb_offset = reg_offset + (slot << 3);
+
+	bcm_qspi_write(qspi, MSPI, msb_offset, swap4bytes(val));
+}
+
+static inline void write_txram_slot_u64(struct bcm_qspi *qspi, int slot,
+					u64 val)
+{
+	u32 reg_offset = MSPI_TXRAM;
+	u32 msb_offset = reg_offset + (slot << 3);
+	u32 lsb_offset = reg_offset + (slot << 3) + 0x4;
+	u32 msb = upper_32_bits(val);
+	u32 lsb = lower_32_bits(val);
+
+	bcm_qspi_write(qspi, MSPI, msb_offset, swap4bytes(msb));
+	bcm_qspi_write(qspi, MSPI, lsb_offset, swap4bytes(lsb));
+}
+
 static inline u32 read_cdram_slot(struct bcm_qspi *qspi, int slot)
 {
 	return bcm_qspi_read(qspi, MSPI, MSPI_CDRAM + (slot << 2));
@ -792,20 +913,43 @@ static int write_to_hw(struct bcm_qspi *qspi, struct spi_device *spi)

 	/* Run until end of transfer or reached the max data */
 	while (!tstatus && slot < MSPI_NUM_CDRAM) {
+		mspi_cdram = MSPI_CDRAM_CONT_BIT;
 		if (tp.trans->bits_per_word <= 8) {
 			const u8 *buf = tp.trans->tx_buf;
 			u8 val = buf ? buf[tp.byte] : 0x00;

 			write_txram_slot_u8(qspi, slot, val);
 			dev_dbg(&qspi->pdev->dev, "WR %02x\n", val);
-		} else {
+		} else if (tp.trans->bits_per_word <= 16) {
 			const u16 *buf = tp.trans->tx_buf;
 			u16 val = buf ? buf[tp.byte / 2] : 0x0000;

 			write_txram_slot_u16(qspi, slot, val);
 			dev_dbg(&qspi->pdev->dev, "WR %04x\n", val);
+		} else if (tp.trans->bits_per_word <= 32) {
+			const u32 *buf = tp.trans->tx_buf;
+			u32 val = buf ? buf[tp.byte/4] : 0x0;
+
+			write_txram_slot_u32(qspi, slot, val);
+			dev_dbg(&qspi->pdev->dev, "WR %08x\n", val);
+		} else if (tp.trans->bits_per_word <= 64) {
+			const u64 *buf = tp.trans->tx_buf;
+			u64 val = (buf ? buf[tp.byte/8] : 0x0);
+
+			/* use the length of delay from SPCR1_LSB */
+			if (bcm_qspi_has_fastbr(qspi))
+				mspi_cdram |= MSPI_CDRAM_DT_BIT;
+
+			write_txram_slot_u64(qspi, slot, val);
+			dev_dbg(&qspi->pdev->dev, "WR %llx\n", val);
 		}
-		mspi_cdram = MSPI_CDRAM_CONT_BIT;
+
+		mspi_cdram |= ((tp.trans->bits_per_word <= 8) ? 0 :
+			       MSPI_CDRAM_BITSE_BIT);
+
+		/* set 3wrire halfduplex mode data from master to slave */
+		if ((spi->mode & SPI_3WIRE) && tp.trans->tx_buf)
+			mspi_cdram |= MSPI_CDRAM_OUTP;

 		if (has_bspi(qspi))
 			mspi_cdram &= ~1;
@ -813,9 +957,6 @@ static int write_to_hw(struct bcm_qspi *qspi, struct spi_device *spi)
 			mspi_cdram |= (~(1 << spi->chip_select) &
 				       MSPI_CDRAM_PCS);

-		mspi_cdram |= ((tp.trans->bits_per_word <= 8) ? 0 :
-				MSPI_CDRAM_BITSE_BIT);
-
 		write_cdram_slot(qspi, slot, mspi_cdram);

 		tstatus = update_qspi_trans_byte_count(qspi, &tp,
@ -1350,7 +1491,8 @@ int bcm_qspi_probe(struct platform_device *pdev,
 	qspi->master = master;

 	master->bus_num = -1;
-	master->mode_bits = SPI_CPHA | SPI_CPOL | SPI_RX_DUAL | SPI_RX_QUAD;
+	master->mode_bits = SPI_CPHA | SPI_CPOL | SPI_RX_DUAL | SPI_RX_QUAD |
+				SPI_3WIRE;
 	master->setup = bcm_qspi_setup;
 	master->transfer_one = bcm_qspi_transfer_one;
 	master->mem_ops = &bcm_qspi_mem_ops;
@ -1460,7 +1602,7 @@ int bcm_qspi_probe(struct platform_device *pdev,
 					       &qspi->dev_ids[val]);
 			if (ret < 0) {
 				dev_err(&pdev->dev, "IRQ %s not found\n", name);
-				goto qspi_probe_err;
+				goto qspi_unprepare_err;
 			}

 			qspi->dev_ids[val].dev = qspi;
@ -1475,7 +1617,7 @@ int bcm_qspi_probe(struct platform_device *pdev,
 	if (!num_ints) {
 		dev_err(&pdev->dev, "no IRQs registered, cannot init driver\n");
 		ret = -EINVAL;
-		goto qspi_probe_err;
+		goto qspi_unprepare_err;
 	}

 	bcm_qspi_hw_init(qspi);
@ -1499,6 +1641,7 @@ int bcm_qspi_probe(struct platform_device *pdev,

 qspi_reg_err:
 	bcm_qspi_hw_uninit(qspi);
+qspi_unprepare_err:
 	clk_disable_unprepare(qspi->clk);
 qspi_probe_err:
 	kfree(qspi->dev_ids);
--- a/drivers/spi/spi-cadence-quadspi.c
+++ b/drivers/spi/spi-cadence-quadspi.c
@ -13,6 +13,7 @@
 #include <linux/dmaengine.h>
 #include <linux/err.h>
 #include <linux/errno.h>
+#include <linux/firmware/xlnx-zynqmp.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/iopoll.h>
@ -35,6 +36,7 @@
 /* Quirks */
 #define CQSPI_NEEDS_WR_DELAY		BIT(0)
 #define CQSPI_DISABLE_DAC_MODE		BIT(1)
+#define CQSPI_SUPPORT_EXTERNAL_DMA	BIT(2)

 /* Capabilities */
 #define CQSPI_SUPPORTS_OCTAL		BIT(0)
@ -82,11 +84,16 @@ struct cqspi_st {
 	u32			wr_delay;
 	bool			use_direct_mode;
 	struct cqspi_flash_pdata f_pdata[CQSPI_MAX_CHIPSELECT];
+	bool			use_dma_read;
+	u32			pd_dev_id;
 };

 struct cqspi_driver_platdata {
 	u32 hwcaps_mask;
 	u8 quirks;
+	int (*indirect_read_dma)(struct cqspi_flash_pdata *f_pdata,
+				 u_char *rxbuf, loff_t from_addr, size_t n_rx);
+	u32 (*get_dma_status)(struct cqspi_st *cqspi);
 };

 /* Operation timeout value */
@ -217,6 +224,8 @@ struct cqspi_driver_platdata {
 #define CQSPI_REG_INDIRECTWRSTARTADDR		0x78
 #define CQSPI_REG_INDIRECTWRBYTES		0x7C

+#define CQSPI_REG_INDTRIG_ADDRRANGE		0x80
+
 #define CQSPI_REG_CMDADDRESS			0x94
 #define CQSPI_REG_CMDREADDATALOWER		0xA0
 #define CQSPI_REG_CMDREADDATAUPPER		0xA4
@ -231,6 +240,23 @@ struct cqspi_driver_platdata {
 #define CQSPI_REG_OP_EXT_WRITE_LSB		16
 #define CQSPI_REG_OP_EXT_STIG_LSB		0

+#define CQSPI_REG_VERSAL_DMA_SRC_ADDR		0x1000
+
+#define CQSPI_REG_VERSAL_DMA_DST_ADDR		0x1800
+#define CQSPI_REG_VERSAL_DMA_DST_SIZE		0x1804
+
+#define CQSPI_REG_VERSAL_DMA_DST_CTRL		0x180C
+
+#define CQSPI_REG_VERSAL_DMA_DST_I_STS		0x1814
+#define CQSPI_REG_VERSAL_DMA_DST_I_EN		0x1818
+#define CQSPI_REG_VERSAL_DMA_DST_I_DIS		0x181C
+#define CQSPI_REG_VERSAL_DMA_DST_DONE_MASK	BIT(1)
+
+#define CQSPI_REG_VERSAL_DMA_DST_ADDR_MSB	0x1828
+
+#define CQSPI_REG_VERSAL_DMA_DST_CTRL_VAL	0xF43FFA00
+#define CQSPI_REG_VERSAL_ADDRRANGE_WIDTH_VAL	0x6
+
 /* Interrupt status bits */
 #define CQSPI_REG_IRQ_MODE_ERR			BIT(0)
 #define CQSPI_REG_IRQ_UNDERFLOW			BIT(1)
@ -250,6 +276,9 @@ struct cqspi_driver_platdata {
 					 CQSPI_REG_IRQ_UNDERFLOW)

 #define CQSPI_IRQ_STATUS_MASK		0x1FFFF
+#define CQSPI_DMA_UNALIGN		0x3
+
+#define CQSPI_REG_VERSAL_DMA_VAL		0x602

 static int cqspi_wait_for_bit(void __iomem *reg, const u32 mask, bool clr)
 {
@ -275,10 +304,26 @@ static u32 cqspi_get_rd_sram_level(struct cqspi_st *cqspi)
 	return reg & CQSPI_REG_SDRAMLEVEL_RD_MASK;
 }

+static u32 cqspi_get_versal_dma_status(struct cqspi_st *cqspi)
+{
+	u32 dma_status;
+
+	dma_status = readl(cqspi->iobase +
+					   CQSPI_REG_VERSAL_DMA_DST_I_STS);
+	writel(dma_status, cqspi->iobase +
+		   CQSPI_REG_VERSAL_DMA_DST_I_STS);
+
+	return dma_status & CQSPI_REG_VERSAL_DMA_DST_DONE_MASK;
+}
+
 static irqreturn_t cqspi_irq_handler(int this_irq, void *dev)
 {
 	struct cqspi_st *cqspi = dev;
 	unsigned int irq_status;
+	struct device *device = &cqspi->pdev->dev;
+	const struct cqspi_driver_platdata *ddata;
+
+	ddata = of_device_get_match_data(device);

 	/* Read interrupt status */
 	irq_status = readl(cqspi->iobase + CQSPI_REG_IRQSTATUS);
@ -286,6 +331,13 @@ static irqreturn_t cqspi_irq_handler(int this_irq, void *dev)
 	/* Clear interrupt */
 	writel(irq_status, cqspi->iobase + CQSPI_REG_IRQSTATUS);

+	if (cqspi->use_dma_read && ddata && ddata->get_dma_status) {
+		if (ddata->get_dma_status(cqspi)) {
+			complete(&cqspi->transfer_complete);
+			return IRQ_HANDLED;
+		}
+	}
+
 	irq_status &= CQSPI_IRQ_MASK_RD | CQSPI_IRQ_MASK_WR;

 	if (irq_status)
@ -781,6 +833,131 @@ failrd:
 	return ret;
 }

+static int cqspi_versal_indirect_read_dma(struct cqspi_flash_pdata *f_pdata,
+					  u_char *rxbuf, loff_t from_addr,
+					  size_t n_rx)
+{
+	struct cqspi_st *cqspi = f_pdata->cqspi;
+	struct device *dev = &cqspi->pdev->dev;
+	void __iomem *reg_base = cqspi->iobase;
+	u32 reg, bytes_to_dma;
+	loff_t addr = from_addr;
+	void *buf = rxbuf;
+	dma_addr_t dma_addr;
+	u8 bytes_rem;
+	int ret = 0;
+
+	bytes_rem = n_rx % 4;
+	bytes_to_dma = (n_rx - bytes_rem);
+
+	if (!bytes_to_dma)
+		goto nondmard;
+
+	ret = zynqmp_pm_ospi_mux_select(cqspi->pd_dev_id, PM_OSPI_MUX_SEL_DMA);
+	if (ret)
+		return ret;
+
+	reg = readl(cqspi->iobase + CQSPI_REG_CONFIG);
+	reg |= CQSPI_REG_CONFIG_DMA_MASK;
+	writel(reg, cqspi->iobase + CQSPI_REG_CONFIG);
+
+	dma_addr = dma_map_single(dev, rxbuf, bytes_to_dma, DMA_FROM_DEVICE);
+	if (dma_mapping_error(dev, dma_addr)) {
+		dev_err(dev, "dma mapping failed\n");
+		return -ENOMEM;
+	}
+
+	writel(from_addr, reg_base + CQSPI_REG_INDIRECTRDSTARTADDR);
+	writel(bytes_to_dma, reg_base + CQSPI_REG_INDIRECTRDBYTES);
+	writel(CQSPI_REG_VERSAL_ADDRRANGE_WIDTH_VAL,
+	       reg_base + CQSPI_REG_INDTRIG_ADDRRANGE);
+
+	/* Clear all interrupts. */
+	writel(CQSPI_IRQ_STATUS_MASK, reg_base + CQSPI_REG_IRQSTATUS);
+
+	/* Enable DMA done interrupt */
+	writel(CQSPI_REG_VERSAL_DMA_DST_DONE_MASK,
+	       reg_base + CQSPI_REG_VERSAL_DMA_DST_I_EN);
+
+	/* Default DMA periph configuration */
+	writel(CQSPI_REG_VERSAL_DMA_VAL, reg_base + CQSPI_REG_DMA);
+
+	/* Configure DMA Dst address */
+	writel(lower_32_bits(dma_addr),
+	       reg_base + CQSPI_REG_VERSAL_DMA_DST_ADDR);
+	writel(upper_32_bits(dma_addr),
+	       reg_base + CQSPI_REG_VERSAL_DMA_DST_ADDR_MSB);
+
+	/* Configure DMA Src address */
+	writel(cqspi->trigger_address, reg_base +
+	       CQSPI_REG_VERSAL_DMA_SRC_ADDR);
+
+	/* Set DMA destination size */
+	writel(bytes_to_dma, reg_base + CQSPI_REG_VERSAL_DMA_DST_SIZE);
+
+	/* Set DMA destination control */
+	writel(CQSPI_REG_VERSAL_DMA_DST_CTRL_VAL,
+	       reg_base + CQSPI_REG_VERSAL_DMA_DST_CTRL);
+
+	writel(CQSPI_REG_INDIRECTRD_START_MASK,
+	       reg_base + CQSPI_REG_INDIRECTRD);
+
+	reinit_completion(&cqspi->transfer_complete);
+
+	if (!wait_for_completion_timeout(&cqspi->transfer_complete,
+					 msecs_to_jiffies(CQSPI_READ_TIMEOUT_MS))) {
+		ret = -ETIMEDOUT;
+		goto failrd;
+	}
+
+	/* Disable DMA interrupt */
+	writel(0x0, cqspi->iobase + CQSPI_REG_VERSAL_DMA_DST_I_DIS);
+
+	/* Clear indirect completion status */
+	writel(CQSPI_REG_INDIRECTRD_DONE_MASK,
+	       cqspi->iobase + CQSPI_REG_INDIRECTRD);
+	dma_unmap_single(dev, dma_addr, bytes_to_dma, DMA_FROM_DEVICE);
+
+	reg = readl(cqspi->iobase + CQSPI_REG_CONFIG);
+	reg &= ~CQSPI_REG_CONFIG_DMA_MASK;
+	writel(reg, cqspi->iobase + CQSPI_REG_CONFIG);
+
+	ret = zynqmp_pm_ospi_mux_select(cqspi->pd_dev_id,
+					PM_OSPI_MUX_SEL_LINEAR);
+	if (ret)
+		return ret;
+
+nondmard:
+	if (bytes_rem) {
+		addr += bytes_to_dma;
+		buf += bytes_to_dma;
+		ret = cqspi_indirect_read_execute(f_pdata, buf, addr,
+						  bytes_rem);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+
+failrd:
+	/* Disable DMA interrupt */
+	writel(0x0, reg_base + CQSPI_REG_VERSAL_DMA_DST_I_DIS);
+
+	/* Cancel the indirect read */
+	writel(CQSPI_REG_INDIRECTWR_CANCEL_MASK,
+	       reg_base + CQSPI_REG_INDIRECTRD);
+
+	dma_unmap_single(dev, dma_addr, bytes_to_dma, DMA_FROM_DEVICE);
+
+	reg = readl(cqspi->iobase + CQSPI_REG_CONFIG);
+	reg &= ~CQSPI_REG_CONFIG_DMA_MASK;
+	writel(reg, cqspi->iobase + CQSPI_REG_CONFIG);
+
+	zynqmp_pm_ospi_mux_select(cqspi->pd_dev_id, PM_OSPI_MUX_SEL_LINEAR);
+
+	return ret;
+}
+
 static int cqspi_write_setup(struct cqspi_flash_pdata *f_pdata,
 			     const struct spi_mem_op *op)
 {
@ -1180,11 +1357,15 @@ static ssize_t cqspi_read(struct cqspi_flash_pdata *f_pdata,
 			  const struct spi_mem_op *op)
 {
 	struct cqspi_st *cqspi = f_pdata->cqspi;
+	struct device *dev = &cqspi->pdev->dev;
+	const struct cqspi_driver_platdata *ddata;
 	loff_t from = op->addr.val;
 	size_t len = op->data.nbytes;
 	u_char *buf = op->data.buf.in;
+	u64 dma_align = (u64)(uintptr_t)buf;
 	int ret;

+	ddata = of_device_get_match_data(dev);
 	ret = cqspi_set_protocol(f_pdata, op);
 	if (ret)
 		return ret;
@ -1196,6 +1377,10 @@ static ssize_t cqspi_read(struct cqspi_flash_pdata *f_pdata,
 	if (cqspi->use_direct_mode && ((from + len) <= cqspi->ahb_size))
 		return cqspi_direct_read_execute(f_pdata, buf, from, len);

+	if (cqspi->use_dma_read && ddata && ddata->indirect_read_dma &&
+	    virt_addr_valid(buf) && ((dma_align & CQSPI_DMA_UNALIGN) == 0))
+		return ddata->indirect_read_dma(f_pdata, buf, from, len);
+
 	return cqspi_indirect_read_execute(f_pdata, buf, from, len);
 }

@ -1299,6 +1484,7 @@ static int cqspi_of_get_pdata(struct cqspi_st *cqspi)
 {
 	struct device *dev = &cqspi->pdev->dev;
 	struct device_node *np = dev->of_node;
+	u32 id[2];

 	cqspi->is_decoded_cs = of_property_read_bool(np, "cdns,is-decoded-cs");

@ -1323,6 +1509,10 @@ static int cqspi_of_get_pdata(struct cqspi_st *cqspi)

 	cqspi->rclk_en = of_property_read_bool(np, "cdns,rclk-en");

+	if (!of_property_read_u32_array(np, "power-domains", id,
+					ARRAY_SIZE(id)))
+		cqspi->pd_dev_id = id[1];
+
 	return 0;
 }

@ -1359,6 +1549,13 @@ static void cqspi_controller_init(struct cqspi_st *cqspi)
 		writel(reg, cqspi->iobase + CQSPI_REG_CONFIG);
 	}

+	/* Enable DMA interface */
+	if (cqspi->use_dma_read) {
+		reg = readl(cqspi->iobase + CQSPI_REG_CONFIG);
+		reg |= CQSPI_REG_CONFIG_DMA_MASK;
+		writel(reg, cqspi->iobase + CQSPI_REG_CONFIG);
+	}
+
 	cqspi_controller_enable(cqspi, 1);
 }

@ -1548,6 +1745,12 @@ static int cqspi_probe(struct platform_device *pdev)
 			master->mode_bits |= SPI_RX_OCTAL | SPI_TX_OCTAL;
 		if (!(ddata->quirks & CQSPI_DISABLE_DAC_MODE))
 			cqspi->use_direct_mode = true;
+		if (ddata->quirks & CQSPI_SUPPORT_EXTERNAL_DMA)
+			cqspi->use_dma_read = true;
+
+		if (of_device_is_compatible(pdev->dev.of_node,
+					    "xlnx,versal-ospi-1.0"))
+			dma_set_mask(&pdev->dev, DMA_BIT_MASK(64));
 	}

 	ret = devm_request_irq(dev, irq, cqspi_irq_handler, 0,
@ -1656,6 +1859,13 @@ static const struct cqspi_driver_platdata intel_lgm_qspi = {
 	.quirks = CQSPI_DISABLE_DAC_MODE,
 };

+static const struct cqspi_driver_platdata versal_ospi = {
+	.hwcaps_mask = CQSPI_SUPPORTS_OCTAL,
+	.quirks = CQSPI_DISABLE_DAC_MODE | CQSPI_SUPPORT_EXTERNAL_DMA,
+	.indirect_read_dma = cqspi_versal_indirect_read_dma,
+	.get_dma_status = cqspi_get_versal_dma_status,
+};
+
 static const struct of_device_id cqspi_dt_ids[] = {
 	{
 		.compatible = "cdns,qspi-nor",
@ -1673,6 +1883,10 @@ static const struct of_device_id cqspi_dt_ids[] = {
 		.compatible = "intel,lgm-qspi",
 		.data = &intel_lgm_qspi,
 	},
+	{
+		.compatible = "xlnx,versal-ospi-1.0",
+		.data = (void *)&versal_ospi,
+	},
 	{ /* end of table */ }
 };

--- a/drivers/spi/spi-cadence-xspi.c
+++ b/drivers/spi/spi-cadence-xspi.c
@ -0,0 +1,642 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Cadence XSPI flash controller driver
+// Copyright (C) 2020-21 Cadence
+
+#include <linux/completion.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/spi-mem.h>
+#include <linux/bitfield.h>
+#include <linux/limits.h>
+#include <linux/log2.h>
+
+#define CDNS_XSPI_MAGIC_NUM_VALUE	0x6522
+#define CDNS_XSPI_MAX_BANKS		8
+#define CDNS_XSPI_NAME			"cadence-xspi"
+
+/*
+ * Note: below are additional auxiliary registers to
+ * configure XSPI controller pin-strap settings
+ */
+
+/* PHY DQ timing register */
+#define CDNS_XSPI_CCP_PHY_DQ_TIMING		0x0000
+
+/* PHY DQS timing register */
+#define CDNS_XSPI_CCP_PHY_DQS_TIMING		0x0004
+
+/* PHY gate loopback control register */
+#define CDNS_XSPI_CCP_PHY_GATE_LPBCK_CTRL	0x0008
+
+/* PHY DLL slave control register */
+#define CDNS_XSPI_CCP_PHY_DLL_SLAVE_CTRL	0x0010
+
+/* DLL PHY control register */
+#define CDNS_XSPI_DLL_PHY_CTRL			0x1034
+
+/* Command registers */
+#define CDNS_XSPI_CMD_REG_0			0x0000
+#define CDNS_XSPI_CMD_REG_1			0x0004
+#define CDNS_XSPI_CMD_REG_2			0x0008
+#define CDNS_XSPI_CMD_REG_3			0x000C
+#define CDNS_XSPI_CMD_REG_4			0x0010
+#define CDNS_XSPI_CMD_REG_5			0x0014
+
+/* Command status registers */
+#define CDNS_XSPI_CMD_STATUS_REG		0x0044
+
+/* Controller status register */
+#define CDNS_XSPI_CTRL_STATUS_REG		0x0100
+#define CDNS_XSPI_INIT_COMPLETED		BIT(16)
+#define CDNS_XSPI_INIT_LEGACY			BIT(9)
+#define CDNS_XSPI_INIT_FAIL			BIT(8)
+#define CDNS_XSPI_CTRL_BUSY			BIT(7)
+
+/* Controller interrupt status register */
+#define CDNS_XSPI_INTR_STATUS_REG		0x0110
+#define CDNS_XSPI_STIG_DONE			BIT(23)
+#define CDNS_XSPI_SDMA_ERROR			BIT(22)
+#define CDNS_XSPI_SDMA_TRIGGER			BIT(21)
+#define CDNS_XSPI_CMD_IGNRD_EN			BIT(20)
+#define CDNS_XSPI_DDMA_TERR_EN			BIT(18)
+#define CDNS_XSPI_CDMA_TREE_EN			BIT(17)
+#define CDNS_XSPI_CTRL_IDLE_EN			BIT(16)
+
+#define CDNS_XSPI_TRD_COMP_INTR_STATUS		0x0120
+#define CDNS_XSPI_TRD_ERR_INTR_STATUS		0x0130
+#define CDNS_XSPI_TRD_ERR_INTR_EN		0x0134
+
+/* Controller interrupt enable register */
+#define CDNS_XSPI_INTR_ENABLE_REG		0x0114
+#define CDNS_XSPI_INTR_EN			BIT(31)
+#define CDNS_XSPI_STIG_DONE_EN			BIT(23)
+#define CDNS_XSPI_SDMA_ERROR_EN			BIT(22)
+#define CDNS_XSPI_SDMA_TRIGGER_EN		BIT(21)
+
+#define CDNS_XSPI_INTR_MASK (CDNS_XSPI_INTR_EN | \
+	CDNS_XSPI_STIG_DONE_EN  | \
+	CDNS_XSPI_SDMA_ERROR_EN | \
+	CDNS_XSPI_SDMA_TRIGGER_EN)
+
+/* Controller config register */
+#define CDNS_XSPI_CTRL_CONFIG_REG		0x0230
+#define CDNS_XSPI_CTRL_WORK_MODE		GENMASK(6, 5)
+
+#define CDNS_XSPI_WORK_MODE_DIRECT		0
+#define CDNS_XSPI_WORK_MODE_STIG		1
+#define CDNS_XSPI_WORK_MODE_ACMD		3
+
+/* SDMA trigger transaction registers */
+#define CDNS_XSPI_SDMA_SIZE_REG			0x0240
+#define CDNS_XSPI_SDMA_TRD_INFO_REG		0x0244
+#define CDNS_XSPI_SDMA_DIR			BIT(8)
+
+/* Controller features register */
+#define CDNS_XSPI_CTRL_FEATURES_REG		0x0F04
+#define CDNS_XSPI_NUM_BANKS			GENMASK(25, 24)
+#define CDNS_XSPI_DMA_DATA_WIDTH		BIT(21)
+#define CDNS_XSPI_NUM_THREADS			GENMASK(3, 0)
+
+/* Controller version register */
+#define CDNS_XSPI_CTRL_VERSION_REG		0x0F00
+#define CDNS_XSPI_MAGIC_NUM			GENMASK(31, 16)
+#define CDNS_XSPI_CTRL_REV			GENMASK(7, 0)
+
+/* STIG Profile 1.0 instruction fields (split into registers) */
+#define CDNS_XSPI_CMD_INSTR_TYPE		GENMASK(6, 0)
+#define CDNS_XSPI_CMD_P1_R1_ADDR0		GENMASK(31, 24)
+#define CDNS_XSPI_CMD_P1_R2_ADDR1		GENMASK(7, 0)
+#define CDNS_XSPI_CMD_P1_R2_ADDR2		GENMASK(15, 8)
+#define CDNS_XSPI_CMD_P1_R2_ADDR3		GENMASK(23, 16)
+#define CDNS_XSPI_CMD_P1_R2_ADDR4		GENMASK(31, 24)
+#define CDNS_XSPI_CMD_P1_R3_ADDR5		GENMASK(7, 0)
+#define CDNS_XSPI_CMD_P1_R3_CMD			GENMASK(23, 16)
+#define CDNS_XSPI_CMD_P1_R3_NUM_ADDR_BYTES	GENMASK(30, 28)
+#define CDNS_XSPI_CMD_P1_R4_ADDR_IOS		GENMASK(1, 0)
+#define CDNS_XSPI_CMD_P1_R4_CMD_IOS		GENMASK(9, 8)
+#define CDNS_XSPI_CMD_P1_R4_BANK		GENMASK(14, 12)
+
+/* STIG data sequence instruction fields (split into registers) */
+#define CDNS_XSPI_CMD_DSEQ_R2_DCNT_L		GENMASK(31, 16)
+#define CDNS_XSPI_CMD_DSEQ_R3_DCNT_H		GENMASK(15, 0)
+#define CDNS_XSPI_CMD_DSEQ_R3_NUM_OF_DUMMY	GENMASK(25, 20)
+#define CDNS_XSPI_CMD_DSEQ_R4_BANK		GENMASK(14, 12)
+#define CDNS_XSPI_CMD_DSEQ_R4_DATA_IOS		GENMASK(9, 8)
+#define CDNS_XSPI_CMD_DSEQ_R4_DIR		BIT(4)
+
+/* STIG command status fields */
+#define CDNS_XSPI_CMD_STATUS_COMPLETED		BIT(15)
+#define CDNS_XSPI_CMD_STATUS_FAILED		BIT(14)
+#define CDNS_XSPI_CMD_STATUS_DQS_ERROR		BIT(3)
+#define CDNS_XSPI_CMD_STATUS_CRC_ERROR		BIT(2)
+#define CDNS_XSPI_CMD_STATUS_BUS_ERROR		BIT(1)
+#define CDNS_XSPI_CMD_STATUS_INV_SEQ_ERROR	BIT(0)
+
+#define CDNS_XSPI_STIG_DONE_FLAG		BIT(0)
+#define CDNS_XSPI_TRD_STATUS			0x0104
+
+/* Helper macros for filling command registers */
+#define CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_1(op, data_phase) ( \
+	FIELD_PREP(CDNS_XSPI_CMD_INSTR_TYPE, (data_phase) ? \
+		CDNS_XSPI_STIG_INSTR_TYPE_1 : CDNS_XSPI_STIG_INSTR_TYPE_0) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R1_ADDR0, (op)->addr.val & 0xff))
+
+#define CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_2(op) ( \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R2_ADDR1, ((op)->addr.val >> 8)  & 0xFF) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R2_ADDR2, ((op)->addr.val >> 16) & 0xFF) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R2_ADDR3, ((op)->addr.val >> 24) & 0xFF) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R2_ADDR4, ((op)->addr.val >> 32) & 0xFF))
+
+#define CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_3(op) ( \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R3_ADDR5, ((op)->addr.val >> 40) & 0xFF) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R3_CMD, (op)->cmd.opcode) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R3_NUM_ADDR_BYTES, (op)->addr.nbytes))
+
+#define CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_4(op, chipsel) ( \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R4_ADDR_IOS, ilog2((op)->addr.buswidth)) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R4_CMD_IOS, ilog2((op)->cmd.buswidth)) | \
+	FIELD_PREP(CDNS_XSPI_CMD_P1_R4_BANK, chipsel))
+
+#define CDNS_XSPI_CMD_FLD_DSEQ_CMD_1(op) \
+	FIELD_PREP(CDNS_XSPI_CMD_INSTR_TYPE, CDNS_XSPI_STIG_INSTR_TYPE_DATA_SEQ)
+
+#define CDNS_XSPI_CMD_FLD_DSEQ_CMD_2(op) \
+	FIELD_PREP(CDNS_XSPI_CMD_DSEQ_R2_DCNT_L, (op)->data.nbytes & 0xFFFF)
+
+#define CDNS_XSPI_CMD_FLD_DSEQ_CMD_3(op) ( \
+	FIELD_PREP(CDNS_XSPI_CMD_DSEQ_R3_DCNT_H, \
+		((op)->data.nbytes >> 16) & 0xffff) | \
+	FIELD_PREP(CDNS_XSPI_CMD_DSEQ_R3_NUM_OF_DUMMY, (op)->dummy.nbytes * 8))
+
+#define CDNS_XSPI_CMD_FLD_DSEQ_CMD_4(op, chipsel) ( \
+	FIELD_PREP(CDNS_XSPI_CMD_DSEQ_R4_BANK, chipsel) | \
+	FIELD_PREP(CDNS_XSPI_CMD_DSEQ_R4_DATA_IOS, \
+		ilog2((op)->data.buswidth)) | \
+	FIELD_PREP(CDNS_XSPI_CMD_DSEQ_R4_DIR, \
+		((op)->data.dir == SPI_MEM_DATA_IN) ? \
+		CDNS_XSPI_STIG_CMD_DIR_READ : CDNS_XSPI_STIG_CMD_DIR_WRITE))
+
+enum cdns_xspi_stig_instr_type {
+	CDNS_XSPI_STIG_INSTR_TYPE_0,
+	CDNS_XSPI_STIG_INSTR_TYPE_1,
+	CDNS_XSPI_STIG_INSTR_TYPE_DATA_SEQ = 127,
+};
+
+enum cdns_xspi_sdma_dir {
+	CDNS_XSPI_SDMA_DIR_READ,
+	CDNS_XSPI_SDMA_DIR_WRITE,
+};
+
+enum cdns_xspi_stig_cmd_dir {
+	CDNS_XSPI_STIG_CMD_DIR_READ,
+	CDNS_XSPI_STIG_CMD_DIR_WRITE,
+};
+
+struct cdns_xspi_dev {
+	struct platform_device *pdev;
+	struct device *dev;
+
+	void __iomem *iobase;
+	void __iomem *auxbase;
+	void __iomem *sdmabase;
+
+	int irq;
+	int cur_cs;
+	unsigned int sdmasize;
+
+	struct completion cmd_complete;
+	struct completion auto_cmd_complete;
+	struct completion sdma_complete;
+	bool sdma_error;
+
+	void *in_buffer;
+	const void *out_buffer;
+
+	u8 hw_num_banks;
+};
+
+static int cdns_xspi_wait_for_controller_idle(struct cdns_xspi_dev *cdns_xspi)
+{
+	u32 ctrl_stat;
+
+	return readl_relaxed_poll_timeout(cdns_xspi->iobase +
+					  CDNS_XSPI_CTRL_STATUS_REG,
+					  ctrl_stat,
+					  ((ctrl_stat &
+					    CDNS_XSPI_CTRL_BUSY) == 0),
+					  100, 1000);
+}
+
+static void cdns_xspi_trigger_command(struct cdns_xspi_dev *cdns_xspi,
+				      u32 cmd_regs[6])
+{
+	writel(cmd_regs[5], cdns_xspi->iobase + CDNS_XSPI_CMD_REG_5);
+	writel(cmd_regs[4], cdns_xspi->iobase + CDNS_XSPI_CMD_REG_4);
+	writel(cmd_regs[3], cdns_xspi->iobase + CDNS_XSPI_CMD_REG_3);
+	writel(cmd_regs[2], cdns_xspi->iobase + CDNS_XSPI_CMD_REG_2);
+	writel(cmd_regs[1], cdns_xspi->iobase + CDNS_XSPI_CMD_REG_1);
+	writel(cmd_regs[0], cdns_xspi->iobase + CDNS_XSPI_CMD_REG_0);
+}
+
+static int cdns_xspi_check_command_status(struct cdns_xspi_dev *cdns_xspi)
+{
+	int ret = 0;
+	u32 cmd_status = readl(cdns_xspi->iobase + CDNS_XSPI_CMD_STATUS_REG);
+
+	if (cmd_status & CDNS_XSPI_CMD_STATUS_COMPLETED) {
+		if ((cmd_status & CDNS_XSPI_CMD_STATUS_FAILED) != 0) {
+			if (cmd_status & CDNS_XSPI_CMD_STATUS_DQS_ERROR) {
+				dev_err(cdns_xspi->dev,
+					"Incorrect DQS pulses detected\n");
+				ret = -EPROTO;
+			}
+			if (cmd_status & CDNS_XSPI_CMD_STATUS_CRC_ERROR) {
+				dev_err(cdns_xspi->dev,
+					"CRC error received\n");
+				ret = -EPROTO;
+			}
+			if (cmd_status & CDNS_XSPI_CMD_STATUS_BUS_ERROR) {
+				dev_err(cdns_xspi->dev,
+					"Error resp on system DMA interface\n");
+				ret = -EPROTO;
+			}
+			if (cmd_status & CDNS_XSPI_CMD_STATUS_INV_SEQ_ERROR) {
+				dev_err(cdns_xspi->dev,
+					"Invalid command sequence detected\n");
+				ret = -EPROTO;
+			}
+		}
+	} else {
+		dev_err(cdns_xspi->dev, "Fatal err - command not completed\n");
+		ret = -EPROTO;
+	}
+
+	return ret;
+}
+
+static void cdns_xspi_set_interrupts(struct cdns_xspi_dev *cdns_xspi,
+				     bool enabled)
+{
+	u32 intr_enable;
+
+	intr_enable = readl(cdns_xspi->iobase + CDNS_XSPI_INTR_ENABLE_REG);
+	if (enabled)
+		intr_enable |= CDNS_XSPI_INTR_MASK;
+	else
+		intr_enable &= ~CDNS_XSPI_INTR_MASK;
+	writel(intr_enable, cdns_xspi->iobase + CDNS_XSPI_INTR_ENABLE_REG);
+}
+
+static int cdns_xspi_controller_init(struct cdns_xspi_dev *cdns_xspi)
+{
+	u32 ctrl_ver;
+	u32 ctrl_features;
+	u16 hw_magic_num;
+
+	ctrl_ver = readl(cdns_xspi->iobase + CDNS_XSPI_CTRL_VERSION_REG);
+	hw_magic_num = FIELD_GET(CDNS_XSPI_MAGIC_NUM, ctrl_ver);
+	if (hw_magic_num != CDNS_XSPI_MAGIC_NUM_VALUE) {
+		dev_err(cdns_xspi->dev,
+			"Incorrect XSPI magic number: %x, expected: %x\n",
+			hw_magic_num, CDNS_XSPI_MAGIC_NUM_VALUE);
+		return -EIO;
+	}
+
+	ctrl_features = readl(cdns_xspi->iobase + CDNS_XSPI_CTRL_FEATURES_REG);
+	cdns_xspi->hw_num_banks = FIELD_GET(CDNS_XSPI_NUM_BANKS, ctrl_features);
+	cdns_xspi_set_interrupts(cdns_xspi, false);
+
+	return 0;
+}
+
+static void cdns_xspi_sdma_handle(struct cdns_xspi_dev *cdns_xspi)
+{
+	u32 sdma_size, sdma_trd_info;
+	u8 sdma_dir;
+
+	sdma_size = readl(cdns_xspi->iobase + CDNS_XSPI_SDMA_SIZE_REG);
+	sdma_trd_info = readl(cdns_xspi->iobase + CDNS_XSPI_SDMA_TRD_INFO_REG);
+	sdma_dir = FIELD_GET(CDNS_XSPI_SDMA_DIR, sdma_trd_info);
+
+	switch (sdma_dir) {
+	case CDNS_XSPI_SDMA_DIR_READ:
+		ioread8_rep(cdns_xspi->sdmabase,
+			    cdns_xspi->in_buffer, sdma_size);
+		break;
+
+	case CDNS_XSPI_SDMA_DIR_WRITE:
+		iowrite8_rep(cdns_xspi->sdmabase,
+			     cdns_xspi->out_buffer, sdma_size);
+		break;
+	}
+}
+
+static int cdns_xspi_send_stig_command(struct cdns_xspi_dev *cdns_xspi,
+				       const struct spi_mem_op *op,
+				       bool data_phase)
+{
+	u32 cmd_regs[6];
+	u32 cmd_status;
+	int ret;
+
+	ret = cdns_xspi_wait_for_controller_idle(cdns_xspi);
+	if (ret < 0)
+		return -EIO;
+
+	writel(FIELD_PREP(CDNS_XSPI_CTRL_WORK_MODE, CDNS_XSPI_WORK_MODE_STIG),
+	       cdns_xspi->iobase + CDNS_XSPI_CTRL_CONFIG_REG);
+
+	cdns_xspi_set_interrupts(cdns_xspi, true);
+	cdns_xspi->sdma_error = false;
+
+	memset(cmd_regs, 0, sizeof(cmd_regs));
+	cmd_regs[1] = CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_1(op, data_phase);
+	cmd_regs[2] = CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_2(op);
+	cmd_regs[3] = CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_3(op);
+	cmd_regs[4] = CDNS_XSPI_CMD_FLD_P1_INSTR_CMD_4(op,
+						       cdns_xspi->cur_cs);
+
+	cdns_xspi_trigger_command(cdns_xspi, cmd_regs);
+
+	if (data_phase) {
+		cmd_regs[0] = CDNS_XSPI_STIG_DONE_FLAG;
+		cmd_regs[1] = CDNS_XSPI_CMD_FLD_DSEQ_CMD_1(op);
+		cmd_regs[2] = CDNS_XSPI_CMD_FLD_DSEQ_CMD_2(op);
+		cmd_regs[3] = CDNS_XSPI_CMD_FLD_DSEQ_CMD_3(op);
+		cmd_regs[4] = CDNS_XSPI_CMD_FLD_DSEQ_CMD_4(op,
+							   cdns_xspi->cur_cs);
+
+		cdns_xspi->in_buffer = op->data.buf.in;
+		cdns_xspi->out_buffer = op->data.buf.out;
+
+		cdns_xspi_trigger_command(cdns_xspi, cmd_regs);
+
+		wait_for_completion(&cdns_xspi->sdma_complete);
+		if (cdns_xspi->sdma_error) {
+			cdns_xspi_set_interrupts(cdns_xspi, false);
+			return -EIO;
+		}
+		cdns_xspi_sdma_handle(cdns_xspi);
+	}
+
+	wait_for_completion(&cdns_xspi->cmd_complete);
+	cdns_xspi_set_interrupts(cdns_xspi, false);
+
+	cmd_status = cdns_xspi_check_command_status(cdns_xspi);
+	if (cmd_status)
+		return -EPROTO;
+
+	return 0;
+}
+
+static int cdns_xspi_mem_op(struct cdns_xspi_dev *cdns_xspi,
+			    struct spi_mem *mem,
+			    const struct spi_mem_op *op)
+{
+	enum spi_mem_data_dir dir = op->data.dir;
+
+	if (cdns_xspi->cur_cs != mem->spi->chip_select)
+		cdns_xspi->cur_cs = mem->spi->chip_select;
+
+	return cdns_xspi_send_stig_command(cdns_xspi, op,
+					   (dir != SPI_MEM_NO_DATA));
+}
+
+static int cdns_xspi_mem_op_execute(struct spi_mem *mem,
+				    const struct spi_mem_op *op)
+{
+	struct cdns_xspi_dev *cdns_xspi =
+		spi_master_get_devdata(mem->spi->master);
+	int ret = 0;
+
+	ret = cdns_xspi_mem_op(cdns_xspi, mem, op);
+
+	return ret;
+}
+
+static int cdns_xspi_adjust_mem_op_size(struct spi_mem *mem, struct spi_mem_op *op)
+{
+	struct cdns_xspi_dev *cdns_xspi =
+		spi_master_get_devdata(mem->spi->master);
+
+	op->data.nbytes = clamp_val(op->data.nbytes, 0, cdns_xspi->sdmasize);
+
+	return 0;
+}
+
+static const struct spi_controller_mem_ops cadence_xspi_mem_ops = {
+	.exec_op = cdns_xspi_mem_op_execute,
+	.adjust_op_size = cdns_xspi_adjust_mem_op_size,
+};
+
+static irqreturn_t cdns_xspi_irq_handler(int this_irq, void *dev)
+{
+	struct cdns_xspi_dev *cdns_xspi = dev;
+	u32 irq_status;
+	irqreturn_t result = IRQ_NONE;
+
+	irq_status = readl(cdns_xspi->iobase + CDNS_XSPI_INTR_STATUS_REG);
+	writel(irq_status, cdns_xspi->iobase + CDNS_XSPI_INTR_STATUS_REG);
+
+	if (irq_status &
+	    (CDNS_XSPI_SDMA_ERROR | CDNS_XSPI_SDMA_TRIGGER |
+	     CDNS_XSPI_STIG_DONE)) {
+		if (irq_status & CDNS_XSPI_SDMA_ERROR) {
+			dev_err(cdns_xspi->dev,
+				"Slave DMA transaction error\n");
+			cdns_xspi->sdma_error = true;
+			complete(&cdns_xspi->sdma_complete);
+		}
+
+		if (irq_status & CDNS_XSPI_SDMA_TRIGGER)
+			complete(&cdns_xspi->sdma_complete);
+
+		if (irq_status & CDNS_XSPI_STIG_DONE)
+			complete(&cdns_xspi->cmd_complete);
+
+		result = IRQ_HANDLED;
+	}
+
+	irq_status = readl(cdns_xspi->iobase + CDNS_XSPI_TRD_COMP_INTR_STATUS);
+	if (irq_status) {
+		writel(irq_status,
+		       cdns_xspi->iobase + CDNS_XSPI_TRD_COMP_INTR_STATUS);
+
+		complete(&cdns_xspi->auto_cmd_complete);
+
+		result = IRQ_HANDLED;
+	}
+
+	return result;
+}
+
+static int cdns_xspi_of_get_plat_data(struct platform_device *pdev)
+{
+	struct device_node *node_prop = pdev->dev.of_node;
+	struct device_node *node_child;
+	unsigned int cs;
+
+	for_each_child_of_node(node_prop, node_child) {
+		if (!of_device_is_available(node_child))
+			continue;
+
+		if (of_property_read_u32(node_child, "reg", &cs)) {
+			dev_err(&pdev->dev, "Couldn't get memory chip select\n");
+			of_node_put(node_child);
+			return -ENXIO;
+		} else if (cs >= CDNS_XSPI_MAX_BANKS) {
+			dev_err(&pdev->dev, "reg (cs) parameter value too large\n");
+			of_node_put(node_child);
+			return -ENXIO;
+		}
+	}
+
+	return 0;
+}
+
+static void cdns_xspi_print_phy_config(struct cdns_xspi_dev *cdns_xspi)
+{
+	struct device *dev = cdns_xspi->dev;
+
+	dev_info(dev, "PHY configuration\n");
+	dev_info(dev, "   * xspi_dll_phy_ctrl: %08x\n",
+		 readl(cdns_xspi->iobase + CDNS_XSPI_DLL_PHY_CTRL));
+	dev_info(dev, "   * phy_dq_timing: %08x\n",
+		 readl(cdns_xspi->auxbase + CDNS_XSPI_CCP_PHY_DQ_TIMING));
+	dev_info(dev, "   * phy_dqs_timing: %08x\n",
+		 readl(cdns_xspi->auxbase + CDNS_XSPI_CCP_PHY_DQS_TIMING));
+	dev_info(dev, "   * phy_gate_loopback_ctrl: %08x\n",
+		 readl(cdns_xspi->auxbase + CDNS_XSPI_CCP_PHY_GATE_LPBCK_CTRL));
+	dev_info(dev, "   * phy_dll_slave_ctrl: %08x\n",
+		 readl(cdns_xspi->auxbase + CDNS_XSPI_CCP_PHY_DLL_SLAVE_CTRL));
+}
+
+static int cdns_xspi_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct spi_master *master = NULL;
+	struct cdns_xspi_dev *cdns_xspi = NULL;
+	struct resource *res;
+	int ret;
+
+	master = devm_spi_alloc_master(dev, sizeof(*cdns_xspi));
+	if (!master)
+		return -ENOMEM;
+
+	master->mode_bits = SPI_3WIRE | SPI_TX_DUAL  | SPI_TX_QUAD  |
+		SPI_RX_DUAL | SPI_RX_QUAD | SPI_TX_OCTAL | SPI_RX_OCTAL |
+		SPI_MODE_0  | SPI_MODE_3;
+
+	master->mem_ops = &cadence_xspi_mem_ops;
+	master->dev.of_node = pdev->dev.of_node;
+	master->bus_num = -1;
+
+	platform_set_drvdata(pdev, master);
+
+	cdns_xspi = spi_master_get_devdata(master);
+	cdns_xspi->pdev = pdev;
+	cdns_xspi->dev = &pdev->dev;
+	cdns_xspi->cur_cs = 0;
+
+	init_completion(&cdns_xspi->cmd_complete);
+	init_completion(&cdns_xspi->auto_cmd_complete);
+	init_completion(&cdns_xspi->sdma_complete);
+
+	ret = cdns_xspi_of_get_plat_data(pdev);
+	if (ret)
+		return -ENODEV;
+
+	cdns_xspi->iobase = devm_platform_ioremap_resource_byname(pdev, "io");
+	if (IS_ERR(cdns_xspi->iobase)) {
+		dev_err(dev, "Failed to remap controller base address\n");
+		return PTR_ERR(cdns_xspi->iobase);
+	}
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "sdma");
+	cdns_xspi->sdmabase = devm_ioremap_resource(dev, res);
+	if (IS_ERR(cdns_xspi->sdmabase)) {
+		dev_err(dev, "Failed to remap SDMA address\n");
+		return PTR_ERR(cdns_xspi->sdmabase);
+	}
+	cdns_xspi->sdmasize = resource_size(res);
+
+	cdns_xspi->auxbase = devm_platform_ioremap_resource_byname(pdev, "aux");
+	if (IS_ERR(cdns_xspi->auxbase)) {
+		dev_err(dev, "Failed to remap AUX address\n");
+		return PTR_ERR(cdns_xspi->auxbase);
+	}
+
+	cdns_xspi->irq = platform_get_irq(pdev, 0);
+	if (cdns_xspi->irq < 0) {
+		dev_err(dev, "Failed to get IRQ\n");
+		return -ENXIO;
+	}
+
+	ret = devm_request_irq(dev, cdns_xspi->irq, cdns_xspi_irq_handler,
+			       IRQF_SHARED, pdev->name, cdns_xspi);
+	if (ret) {
+		dev_err(dev, "Failed to request IRQ: %d\n", cdns_xspi->irq);
+		return ret;
+	}
+
+	cdns_xspi_print_phy_config(cdns_xspi);
+
+	ret = cdns_xspi_controller_init(cdns_xspi);
+	if (ret) {
+		dev_err(dev, "Failed to initialize controller\n");
+		return ret;
+	}
+
+	master->num_chipselect = 1 << cdns_xspi->hw_num_banks;
+
+	ret = devm_spi_register_master(dev, master);
+	if (ret) {
+		dev_err(dev, "Failed to register SPI master\n");
+		return ret;
+	}
+
+	dev_info(dev, "Successfully registered SPI master\n");
+
+	return 0;
+}
+
+#ifdef CONFIG_OF
+static const struct of_device_id cdns_xspi_of_match[] = {
+	{
+		.compatible = "cdns,xspi-nor",
+	},
+	{ /* end of table */}
+};
+MODULE_DEVICE_TABLE(of, cdns_xspi_of_match);
+#else
+#define cdns_xspi_of_match NULL
+#endif /* CONFIG_OF */
+
+static struct platform_driver cdns_xspi_platform_driver = {
+	.probe          = cdns_xspi_probe,
+	.remove         = NULL,
+	.driver = {
+		.name = CDNS_XSPI_NAME,
+		.of_match_table = cdns_xspi_of_match,
+	},
+};
+
+module_platform_driver(cdns_xspi_platform_driver);
+
+MODULE_DESCRIPTION("Cadence XSPI Controller Driver");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:" CDNS_XSPI_NAME);
+MODULE_AUTHOR("Konrad Kociolek <konrad@cadence.com>");
+MODULE_AUTHOR("Jayshri Pawar <jpawar@cadence.com>");
+MODULE_AUTHOR("Parshuram Thombare <pthombar@cadence.com>");
--- a/drivers/spi/spi-fsi.c
+++ b/drivers/spi/spi-fsi.c
@ -67,9 +67,14 @@
 	 SPI_FSI_STATUS_RDR_OVERRUN)
 #define SPI_FSI_PORT_CTRL		0x9

+struct fsi2spi {
+	struct fsi_device *fsi; /* FSI2SPI CFAM engine device */
+	struct mutex lock; /* lock access to the device */
+};
+
 struct fsi_spi {
 	struct device *dev;	/* SPI controller device */
-	struct fsi_device *fsi;	/* FSI2SPI CFAM engine device */
+	struct fsi2spi *bridge; /* FSI2SPI device */
 	u32 base;
 };

@ -104,7 +109,7 @@ static int fsi_spi_check_status(struct fsi_spi *ctx)
 	u32 sts;
 	__be32 sts_be;

-	rc = fsi_device_read(ctx->fsi, FSI2SPI_STATUS, &sts_be,
+	rc = fsi_device_read(ctx->bridge->fsi, FSI2SPI_STATUS, &sts_be,
 			     sizeof(sts_be));
 	if (rc)
 		return rc;
@ -120,73 +125,91 @@ static int fsi_spi_check_status(struct fsi_spi *ctx)

 static int fsi_spi_read_reg(struct fsi_spi *ctx, u32 offset, u64 *value)
 {
-	int rc;
+	int rc = 0;
 	__be32 cmd_be;
 	__be32 data_be;
 	u32 cmd = offset + ctx->base;
+	struct fsi2spi *bridge = ctx->bridge;

 	*value = 0ULL;

 	if (cmd & FSI2SPI_CMD_WRITE)
 		return -EINVAL;

-	cmd_be = cpu_to_be32(cmd);
-	rc = fsi_device_write(ctx->fsi, FSI2SPI_CMD, &cmd_be, sizeof(cmd_be));
+	rc = mutex_lock_interruptible(&bridge->lock);
 	if (rc)
 		return rc;

+	cmd_be = cpu_to_be32(cmd);
+	rc = fsi_device_write(bridge->fsi, FSI2SPI_CMD, &cmd_be,
+			      sizeof(cmd_be));
+	if (rc)
+		goto unlock;
+
 	rc = fsi_spi_check_status(ctx);
 	if (rc)
-		return rc;
+		goto unlock;

-	rc = fsi_device_read(ctx->fsi, FSI2SPI_DATA0, &data_be,
+	rc = fsi_device_read(bridge->fsi, FSI2SPI_DATA0, &data_be,
 			     sizeof(data_be));
 	if (rc)
-		return rc;
+		goto unlock;

 	*value |= (u64)be32_to_cpu(data_be) << 32;

-	rc = fsi_device_read(ctx->fsi, FSI2SPI_DATA1, &data_be,
+	rc = fsi_device_read(bridge->fsi, FSI2SPI_DATA1, &data_be,
 			     sizeof(data_be));
 	if (rc)
-		return rc;
+		goto unlock;

 	*value |= (u64)be32_to_cpu(data_be);
 	dev_dbg(ctx->dev, "Read %02x[%016llx].\n", offset, *value);

-	return 0;
+unlock:
+	mutex_unlock(&bridge->lock);
+	return rc;
 }

 static int fsi_spi_write_reg(struct fsi_spi *ctx, u32 offset, u64 value)
 {
-	int rc;
+	int rc = 0;
 	__be32 cmd_be;
 	__be32 data_be;
 	u32 cmd = offset + ctx->base;
+	struct fsi2spi *bridge = ctx->bridge;

 	if (cmd & FSI2SPI_CMD_WRITE)
 		return -EINVAL;

+	rc = mutex_lock_interruptible(&bridge->lock);
+	if (rc)
+		return rc;
+
 	dev_dbg(ctx->dev, "Write %02x[%016llx].\n", offset, value);

 	data_be = cpu_to_be32(upper_32_bits(value));
-	rc = fsi_device_write(ctx->fsi, FSI2SPI_DATA0, &data_be,
+	rc = fsi_device_write(bridge->fsi, FSI2SPI_DATA0, &data_be,
 			      sizeof(data_be));
 	if (rc)
-		return rc;
+		goto unlock;

 	data_be = cpu_to_be32(lower_32_bits(value));
-	rc = fsi_device_write(ctx->fsi, FSI2SPI_DATA1, &data_be,
+	rc = fsi_device_write(bridge->fsi, FSI2SPI_DATA1, &data_be,
 			      sizeof(data_be));
 	if (rc)
-		return rc;
+		goto unlock;

 	cmd_be = cpu_to_be32(cmd | FSI2SPI_CMD_WRITE);
-	rc = fsi_device_write(ctx->fsi, FSI2SPI_CMD, &cmd_be, sizeof(cmd_be));
+	rc = fsi_device_write(bridge->fsi, FSI2SPI_CMD, &cmd_be,
+			      sizeof(cmd_be));
 	if (rc)
-		return rc;
+		goto unlock;

-	return fsi_spi_check_status(ctx);
+	rc = fsi_spi_check_status(ctx);
+
+unlock:
+	mutex_unlock(&bridge->lock);
+	return rc;
 }

 static int fsi_spi_data_in(u64 in, u8 *rx, int len)
@ -234,6 +257,26 @@ static int fsi_spi_reset(struct fsi_spi *ctx)
 	return fsi_spi_write_reg(ctx, SPI_FSI_STATUS, 0ULL);
 }

+static int fsi_spi_status(struct fsi_spi *ctx, u64 *status, const char *dir)
+{
+	int rc = fsi_spi_read_reg(ctx, SPI_FSI_STATUS, status);
+
+	if (rc)
+		return rc;
+
+	if (*status & SPI_FSI_STATUS_ANY_ERROR) {
+		dev_err(ctx->dev, "%s error: %016llx\n", dir, *status);
+
+		rc = fsi_spi_reset(ctx);
+		if (rc)
+			return rc;
+
+		return -EREMOTEIO;
+	}
+
+	return 0;
+}
+
 static void fsi_spi_sequence_add(struct fsi_spi_sequence *seq, u8 val)
 {
 	/*
@ -273,18 +316,9 @@ static int fsi_spi_transfer_data(struct fsi_spi *ctx,
 				return rc;

 			do {
-				rc = fsi_spi_read_reg(ctx, SPI_FSI_STATUS,
-						      &status);
+				rc = fsi_spi_status(ctx, &status, "TX");
 				if (rc)
 					return rc;
-
-				if (status & SPI_FSI_STATUS_ANY_ERROR) {
-					rc = fsi_spi_reset(ctx);
-					if (rc)
-						return rc;
-
-					return -EREMOTEIO;
-				}
 			} while (status & SPI_FSI_STATUS_TDR_FULL);

 			sent += nb;
@ -296,18 +330,9 @@ static int fsi_spi_transfer_data(struct fsi_spi *ctx,

 		while (transfer->len > recv) {
 			do {
-				rc = fsi_spi_read_reg(ctx, SPI_FSI_STATUS,
-						      &status);
+				rc = fsi_spi_status(ctx, &status, "RX");
 				if (rc)
 					return rc;
-
-				if (status & SPI_FSI_STATUS_ANY_ERROR) {
-					rc = fsi_spi_reset(ctx);
-					if (rc)
-						return rc;
-
-					return -EREMOTEIO;
-				}
 			} while (!(status & SPI_FSI_STATUS_RDR_FULL));

 			rc = fsi_spi_read_reg(ctx, SPI_FSI_DATA_RX, &in);
@ -348,8 +373,12 @@ static int fsi_spi_transfer_init(struct fsi_spi *ctx)
 		if (status & (SPI_FSI_STATUS_ANY_ERROR |
 			      SPI_FSI_STATUS_TDR_FULL |
 			      SPI_FSI_STATUS_RDR_FULL)) {
-			if (reset)
+			if (reset) {
+				dev_err(ctx->dev,
+					"Initialization error: %08llx\n",
+					status);
 				return -EIO;
+			}

 			rc = fsi_spi_reset(ctx);
 			if (rc)
@ -388,7 +417,7 @@ static int fsi_spi_transfer_one_message(struct spi_controller *ctlr,
 	struct spi_transfer *transfer;
 	struct fsi_spi *ctx = spi_controller_get_devdata(ctlr);

-	rc = fsi_spi_check_mux(ctx->fsi, ctx->dev);
+	rc = fsi_spi_check_mux(ctx->bridge->fsi, ctx->dev);
 	if (rc)
 		goto error;

@ -478,12 +507,20 @@ static int fsi_spi_probe(struct device *dev)
 	int rc;
 	struct device_node *np;
 	int num_controllers_registered = 0;
+	struct fsi2spi *bridge;
 	struct fsi_device *fsi = to_fsi_dev(dev);

 	rc = fsi_spi_check_mux(fsi, dev);
 	if (rc)
 		return -ENODEV;

+	bridge = devm_kzalloc(dev, sizeof(*bridge), GFP_KERNEL);
+	if (!bridge)
+		return -ENOMEM;
+
+	bridge->fsi = fsi;
+	mutex_init(&bridge->lock);
+
 	for_each_available_child_of_node(dev->of_node, np) {
 		u32 base;
 		struct fsi_spi *ctx;
@ -506,7 +543,7 @@ static int fsi_spi_probe(struct device *dev)

 		ctx = spi_controller_get_devdata(ctlr);
 		ctx->dev = &ctlr->dev;
-		ctx->fsi = fsi;
+		ctx->bridge = bridge;
 		ctx->base = base + SPI_FSI_BASE;

 		rc = devm_spi_register_controller(dev, ctlr);
--- a/drivers/spi/spi-geni-qcom.c
+++ b/drivers/spi/spi-geni-qcom.c
@ -2,6 +2,9 @@
 // Copyright (c) 2017-2018, The Linux foundation. All rights reserved.

 #include <linux/clk.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/dma/qcom-gpi-dma.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/log2.h>
@ -63,6 +66,15 @@
 #define TIMESTAMP_AFTER		BIT(3)
 #define POST_CMD_DELAY		BIT(4)

+#define GSI_LOOPBACK_EN		BIT(0)
+#define GSI_CS_TOGGLE		BIT(3)
+#define GSI_CPHA		BIT(4)
+#define GSI_CPOL		BIT(5)
+
+#define MAX_TX_SG		3
+#define NUM_SPI_XFER		8
+#define SPI_XFER_TIMEOUT_MS	250
+
 struct spi_geni_master {
 	struct geni_se se;
 	struct device *dev;
@ -84,6 +96,9 @@ struct spi_geni_master {
 	int irq;
 	bool cs_flag;
 	bool abort_failed;
+	struct dma_chan *tx;
+	struct dma_chan *rx;
+	int cur_xfer_mode;
 };

 static int get_spi_clk_cfg(unsigned int speed_hz,
@ -330,34 +345,197 @@ static int setup_fifo_params(struct spi_device *spi_slv,
 	return geni_spi_set_clock_and_bw(mas, spi_slv->max_speed_hz);
 }

+static void
+spi_gsi_callback_result(void *cb, const struct dmaengine_result *result)
+{
+	struct spi_master *spi = cb;
+
+	if (result->result != DMA_TRANS_NOERROR) {
+		dev_err(&spi->dev, "DMA txn failed: %d\n", result->result);
+		return;
+	}
+
+	if (!result->residue) {
+		dev_dbg(&spi->dev, "DMA txn completed\n");
+		spi_finalize_current_transfer(spi);
+	} else {
+		dev_err(&spi->dev, "DMA xfer has pending: %d\n", result->residue);
+	}
+}
+
+static int setup_gsi_xfer(struct spi_transfer *xfer, struct spi_geni_master *mas,
+			  struct spi_device *spi_slv, struct spi_master *spi)
+{
+	unsigned long flags = DMA_PREP_INTERRUPT | DMA_CTRL_ACK;
+	struct dma_slave_config config = {};
+	struct gpi_spi_config peripheral = {};
+	struct dma_async_tx_descriptor *tx_desc, *rx_desc;
+	int ret;
+
+	config.peripheral_config = &peripheral;
+	config.peripheral_size = sizeof(peripheral);
+	peripheral.set_config = true;
+
+	if (xfer->bits_per_word != mas->cur_bits_per_word ||
+	    xfer->speed_hz != mas->cur_speed_hz) {
+		mas->cur_bits_per_word = xfer->bits_per_word;
+		mas->cur_speed_hz = xfer->speed_hz;
+	}
+
+	if (xfer->tx_buf && xfer->rx_buf) {
+		peripheral.cmd = SPI_DUPLEX;
+	} else if (xfer->tx_buf) {
+		peripheral.cmd = SPI_TX;
+		peripheral.rx_len = 0;
+	} else if (xfer->rx_buf) {
+		peripheral.cmd = SPI_RX;
+		if (!(mas->cur_bits_per_word % MIN_WORD_LEN)) {
+			peripheral.rx_len = ((xfer->len << 3) / mas->cur_bits_per_word);
+		} else {
+			int bytes_per_word = (mas->cur_bits_per_word / BITS_PER_BYTE) + 1;
+
+			peripheral.rx_len = (xfer->len / bytes_per_word);
+		}
+	}
+
+	peripheral.loopback_en = !!(spi_slv->mode & SPI_LOOP);
+	peripheral.clock_pol_high = !!(spi_slv->mode & SPI_CPOL);
+	peripheral.data_pol_high = !!(spi_slv->mode & SPI_CPHA);
+	peripheral.cs = spi_slv->chip_select;
+	peripheral.pack_en = true;
+	peripheral.word_len = xfer->bits_per_word - MIN_WORD_LEN;
+
+	ret = get_spi_clk_cfg(mas->cur_speed_hz, mas,
+			      &peripheral.clk_src, &peripheral.clk_div);
+	if (ret) {
+		dev_err(mas->dev, "Err in get_spi_clk_cfg() :%d\n", ret);
+		return ret;
+	}
+
+	if (!xfer->cs_change) {
+		if (!list_is_last(&xfer->transfer_list, &spi->cur_msg->transfers))
+			peripheral.fragmentation = FRAGMENTATION;
+	}
+
+	if (peripheral.cmd & SPI_RX) {
+		dmaengine_slave_config(mas->rx, &config);
+		rx_desc = dmaengine_prep_slave_sg(mas->rx, xfer->rx_sg.sgl, xfer->rx_sg.nents,
+						  DMA_DEV_TO_MEM, flags);
+		if (!rx_desc) {
+			dev_err(mas->dev, "Err setting up rx desc\n");
+			return -EIO;
+		}
+	}
+
+	/*
+	 * Prepare the TX always, even for RX or tx_buf being null, we would
+	 * need TX to be prepared per GSI spec
+	 */
+	dmaengine_slave_config(mas->tx, &config);
+	tx_desc = dmaengine_prep_slave_sg(mas->tx, xfer->tx_sg.sgl, xfer->tx_sg.nents,
+					  DMA_MEM_TO_DEV, flags);
+	if (!tx_desc) {
+		dev_err(mas->dev, "Err setting up tx desc\n");
+		return -EIO;
+	}
+
+	tx_desc->callback_result = spi_gsi_callback_result;
+	tx_desc->callback_param = spi;
+
+	if (peripheral.cmd & SPI_RX)
+		dmaengine_submit(rx_desc);
+	dmaengine_submit(tx_desc);
+
+	if (peripheral.cmd & SPI_RX)
+		dma_async_issue_pending(mas->rx);
+
+	dma_async_issue_pending(mas->tx);
+	return 1;
+}
+
+static bool geni_can_dma(struct spi_controller *ctlr,
+			 struct spi_device *slv, struct spi_transfer *xfer)
+{
+	struct spi_geni_master *mas = spi_master_get_devdata(slv->master);
+
+	/* check if dma is supported */
+	return mas->cur_xfer_mode != GENI_SE_FIFO;
+}
+
 static int spi_geni_prepare_message(struct spi_master *spi,
 					struct spi_message *spi_msg)
 {
-	int ret;
 	struct spi_geni_master *mas = spi_master_get_devdata(spi);
+	int ret;

+	switch (mas->cur_xfer_mode) {
+	case GENI_SE_FIFO:
 		if (spi_geni_is_abort_still_pending(mas))
 			return -EBUSY;
-
 		ret = setup_fifo_params(spi_msg->spi, spi);
 		if (ret)
 			dev_err(mas->dev, "Couldn't select mode %d\n", ret);
 		return ret;
+
+	case GENI_GPI_DMA:
+		/* nothing to do for GPI DMA */
+		return 0;
+	}
+
+	dev_err(mas->dev, "Mode not supported %d", mas->cur_xfer_mode);
+	return -EINVAL;
+}
+
+static int spi_geni_grab_gpi_chan(struct spi_geni_master *mas)
+{
+	int ret;
+
+	mas->tx = dma_request_chan(mas->dev, "tx");
+	ret = dev_err_probe(mas->dev, IS_ERR(mas->tx), "Failed to get tx DMA ch\n");
+	if (ret < 0)
+		goto err_tx;
+
+	mas->rx = dma_request_chan(mas->dev, "rx");
+	ret = dev_err_probe(mas->dev, IS_ERR(mas->rx), "Failed to get rx DMA ch\n");
+	if (ret < 0)
+		goto err_rx;
+
+	return 0;
+
+err_rx:
+	dma_release_channel(mas->tx);
+	mas->tx = NULL;
+err_tx:
+	mas->rx = NULL;
+	return ret;
+}
+
+static void spi_geni_release_dma_chan(struct spi_geni_master *mas)
+{
+	if (mas->rx) {
+		dma_release_channel(mas->rx);
+		mas->rx = NULL;
+	}
+
+	if (mas->tx) {
+		dma_release_channel(mas->tx);
+		mas->tx = NULL;
+	}
 }

 static int spi_geni_init(struct spi_geni_master *mas)
 {
 	struct geni_se *se = &mas->se;
 	unsigned int proto, major, minor, ver;
-	u32 spi_tx_cfg;
+	u32 spi_tx_cfg, fifo_disable;
+	int ret = -ENXIO;

 	pm_runtime_get_sync(mas->dev);

 	proto = geni_se_read_proto(se);
 	if (proto != GENI_SE_SPI) {
 		dev_err(mas->dev, "Invalid proto %d\n", proto);
-		pm_runtime_put(mas->dev);
-		return -ENXIO;
+		goto out_pm;
 	}
 	mas->tx_fifo_depth = geni_se_get_tx_fifo_depth(se);

@ -380,15 +558,38 @@ static int spi_geni_init(struct spi_geni_master *mas)
 	else
 		mas->oversampling = 1;

+	fifo_disable = readl(se->base + GENI_IF_DISABLE_RO) & FIFO_IF_DISABLE;
+	switch (fifo_disable) {
+	case 1:
+		ret = spi_geni_grab_gpi_chan(mas);
+		if (!ret) { /* success case */
+			mas->cur_xfer_mode = GENI_GPI_DMA;
+			geni_se_select_mode(se, GENI_GPI_DMA);
+			dev_dbg(mas->dev, "Using GPI DMA mode for SPI\n");
+			break;
+		}
+		/*
+		 * in case of failure to get dma channel, we can still do the
+		 * FIFO mode, so fallthrough
+		 */
+		dev_warn(mas->dev, "FIFO mode disabled, but couldn't get DMA, fall back to FIFO mode\n");
+		fallthrough;
+
+	case 0:
+		mas->cur_xfer_mode = GENI_SE_FIFO;
 		geni_se_select_mode(se, GENI_SE_FIFO);
+		ret = 0;
+		break;
+	}

 	/* We always control CS manually */
 	spi_tx_cfg = readl(se->base + SE_SPI_TRANS_CFG);
 	spi_tx_cfg &= ~CS_TOGGLE;
 	writel(spi_tx_cfg, se->base + SE_SPI_TRANS_CFG);

+out_pm:
 	pm_runtime_put(mas->dev);
-	return 0;
+	return ret;
 }

 static unsigned int geni_byte_per_fifo_word(struct spi_geni_master *mas)
@ -569,9 +770,12 @@ static int spi_geni_transfer_one(struct spi_master *spi,
 	if (!xfer->len)
 		return 0;

+	if (mas->cur_xfer_mode == GENI_SE_FIFO) {
 		setup_fifo_xfer(xfer, mas, slv->mode, spi);
 		return 1;
 	}
+	return setup_gsi_xfer(xfer, mas, slv, spi);
+}

 static irqreturn_t geni_spi_isr(int irq, void *data)
 {
@ -665,6 +869,13 @@ static int spi_geni_probe(struct platform_device *pdev)
 	if (irq < 0)
 		return irq;

+	ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64));
+	if (ret) {
+		ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32));
+		if (ret)
+			return dev_err_probe(dev, ret, "could not set DMA mask\n");
+	}
+
 	base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(base))
 		return PTR_ERR(base);
@ -704,9 +915,10 @@ static int spi_geni_probe(struct platform_device *pdev)
 	spi->max_speed_hz = 50000000;
 	spi->prepare_message = spi_geni_prepare_message;
 	spi->transfer_one = spi_geni_transfer_one;
+	spi->can_dma = geni_can_dma;
+	spi->dma_map_dev = dev->parent;
 	spi->auto_runtime_pm = true;
 	spi->handle_err = handle_fifo_timeout;
-	spi->set_cs = spi_geni_set_cs;
 	spi->use_gpio_descriptors = true;

 	init_completion(&mas->cs_done);
@ -732,9 +944,17 @@ static int spi_geni_probe(struct platform_device *pdev)
 	if (ret)
 		goto spi_geni_probe_runtime_disable;

+	/*
+	 * check the mode supported and set_cs for fifo mode only
+	 * for dma (gsi) mode, the gsi will set cs based on params passed in
+	 * TRE
+	 */
+	if (mas->cur_xfer_mode == GENI_SE_FIFO)
+		spi->set_cs = spi_geni_set_cs;
+
 	ret = request_irq(mas->irq, geni_spi_isr, 0, dev_name(dev), spi);
 	if (ret)
-		goto spi_geni_probe_runtime_disable;
+		goto spi_geni_release_dma;

 	ret = spi_register_master(spi);
 	if (ret)
@ -743,6 +963,8 @@ static int spi_geni_probe(struct platform_device *pdev)
 	return 0;
 spi_geni_probe_free_irq:
 	free_irq(mas->irq, spi);
+spi_geni_release_dma:
+	spi_geni_release_dma_chan(mas);
 spi_geni_probe_runtime_disable:
 	pm_runtime_disable(dev);
 	return ret;
@ -756,6 +978,8 @@ static int spi_geni_remove(struct platform_device *pdev)
 	/* Unregister _before_ disabling pm_runtime() so we stop transfers */
 	spi_unregister_master(spi);

+	spi_geni_release_dma_chan(mas);
+
 	free_irq(mas->irq, spi);
 	pm_runtime_disable(&pdev->dev);
 	return 0;
--- a/drivers/spi/spi-ingenic.c
+++ b/drivers/spi/spi-ingenic.c
@ -0,0 +1,482 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * SPI bus driver for the Ingenic JZ47xx SoCs
+ * Copyright (c) 2017-2021 Artur Rojek <contact@artur-rojek.eu>
+ * Copyright (c) 2017-2021 Paul Cercueil <paul@crapouillou.net>
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/iopoll.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/spi/spi.h>
+
+#define REG_SSIDR	0x0
+#define REG_SSICR0	0x4
+#define REG_SSICR1	0x8
+#define REG_SSISR	0xc
+#define REG_SSIGR	0x18
+
+#define REG_SSICR0_TENDIAN_LSB		BIT(19)
+#define REG_SSICR0_RENDIAN_LSB		BIT(17)
+#define REG_SSICR0_SSIE			BIT(15)
+#define REG_SSICR0_LOOP			BIT(10)
+#define REG_SSICR0_EACLRUN		BIT(7)
+#define REG_SSICR0_FSEL			BIT(6)
+#define REG_SSICR0_TFLUSH		BIT(2)
+#define REG_SSICR0_RFLUSH		BIT(1)
+
+#define REG_SSICR1_FRMHL_MASK		(BIT(31) | BIT(30))
+#define REG_SSICR1_FRMHL		BIT(30)
+#define REG_SSICR1_LFST			BIT(25)
+#define REG_SSICR1_UNFIN		BIT(23)
+#define REG_SSICR1_PHA			BIT(1)
+#define REG_SSICR1_POL			BIT(0)
+
+#define REG_SSISR_END			BIT(7)
+#define REG_SSISR_BUSY			BIT(6)
+#define REG_SSISR_TFF			BIT(5)
+#define REG_SSISR_RFE			BIT(4)
+#define REG_SSISR_RFHF			BIT(2)
+#define REG_SSISR_UNDR			BIT(1)
+#define REG_SSISR_OVER			BIT(0)
+
+#define SPI_INGENIC_FIFO_SIZE		128u
+
+struct jz_soc_info {
+	u32 bits_per_word_mask;
+	struct reg_field flen_field;
+	bool has_trendian;
+};
+
+struct ingenic_spi {
+	const struct jz_soc_info *soc_info;
+	struct clk *clk;
+	struct resource *mem_res;
+
+	struct regmap *map;
+	struct regmap_field *flen_field;
+};
+
+static int spi_ingenic_wait(struct ingenic_spi *priv,
+			    unsigned long mask,
+			    bool condition)
+{
+	unsigned int val;
+
+	return regmap_read_poll_timeout(priv->map, REG_SSISR, val,
+					!!(val & mask) == condition,
+					100, 10000);
+}
+
+static void spi_ingenic_set_cs(struct spi_device *spi, bool disable)
+{
+	struct ingenic_spi *priv = spi_controller_get_devdata(spi->controller);
+
+	if (disable) {
+		regmap_clear_bits(priv->map, REG_SSICR1, REG_SSICR1_UNFIN);
+		regmap_clear_bits(priv->map, REG_SSISR,
+				  REG_SSISR_UNDR | REG_SSISR_OVER);
+
+		spi_ingenic_wait(priv, REG_SSISR_END, true);
+	} else {
+		regmap_set_bits(priv->map, REG_SSICR1, REG_SSICR1_UNFIN);
+	}
+
+	regmap_set_bits(priv->map, REG_SSICR0,
+			REG_SSICR0_RFLUSH | REG_SSICR0_TFLUSH);
+}
+
+static void spi_ingenic_prepare_transfer(struct ingenic_spi *priv,
+					 struct spi_device *spi,
+					 struct spi_transfer *xfer)
+{
+	unsigned long clk_hz = clk_get_rate(priv->clk);
+	u32 cdiv, speed_hz = xfer->speed_hz ?: spi->max_speed_hz,
+	    bits_per_word = xfer->bits_per_word ?: spi->bits_per_word;
+
+	cdiv = clk_hz / (speed_hz * 2);
+	cdiv = clamp(cdiv, 1u, 0x100u) - 1;
+
+	regmap_write(priv->map, REG_SSIGR, cdiv);
+
+	regmap_field_write(priv->flen_field, bits_per_word - 2);
+}
+
+static void spi_ingenic_finalize_transfer(void *controller)
+{
+	spi_finalize_current_transfer(controller);
+}
+
+static struct dma_async_tx_descriptor *
+spi_ingenic_prepare_dma(struct spi_controller *ctlr, struct dma_chan *chan,
+			struct sg_table *sg, enum dma_transfer_direction dir,
+			unsigned int bits)
+{
+	struct ingenic_spi *priv = spi_controller_get_devdata(ctlr);
+	struct dma_slave_config cfg = {
+		.direction = dir,
+		.src_addr = priv->mem_res->start + REG_SSIDR,
+		.dst_addr = priv->mem_res->start + REG_SSIDR,
+	};
+	struct dma_async_tx_descriptor *desc;
+	dma_cookie_t cookie;
+	int ret;
+
+	if (bits > 16) {
+		cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		cfg.src_maxburst = cfg.dst_maxburst = 4;
+	} else if (bits > 8) {
+		cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		cfg.src_maxburst = cfg.dst_maxburst = 2;
+	} else {
+		cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		cfg.src_maxburst = cfg.dst_maxburst = 1;
+	}
+
+	ret = dmaengine_slave_config(chan, &cfg);
+	if (ret)
+		return ERR_PTR(ret);
+
+	desc = dmaengine_prep_slave_sg(chan, sg->sgl, sg->nents, dir,
+				       DMA_PREP_INTERRUPT);
+	if (!desc)
+		return ERR_PTR(-ENOMEM);
+
+	if (dir == DMA_DEV_TO_MEM) {
+		desc->callback = spi_ingenic_finalize_transfer;
+		desc->callback_param = ctlr;
+	}
+
+	cookie = dmaengine_submit(desc);
+
+	ret = dma_submit_error(cookie);
+	if (ret) {
+		dmaengine_desc_free(desc);
+		return ERR_PTR(ret);
+	}
+
+	return desc;
+}
+
+static int spi_ingenic_dma_tx(struct spi_controller *ctlr,
+			      struct spi_transfer *xfer, unsigned int bits)
+{
+	struct dma_async_tx_descriptor *rx_desc, *tx_desc;
+
+	rx_desc = spi_ingenic_prepare_dma(ctlr, ctlr->dma_rx,
+					  &xfer->rx_sg, DMA_DEV_TO_MEM, bits);
+	if (IS_ERR(rx_desc))
+		return PTR_ERR(rx_desc);
+
+	tx_desc = spi_ingenic_prepare_dma(ctlr, ctlr->dma_tx,
+					  &xfer->tx_sg, DMA_MEM_TO_DEV, bits);
+	if (IS_ERR(tx_desc)) {
+		dmaengine_terminate_async(ctlr->dma_rx);
+		dmaengine_desc_free(rx_desc);
+		return PTR_ERR(tx_desc);
+	}
+
+	dma_async_issue_pending(ctlr->dma_rx);
+	dma_async_issue_pending(ctlr->dma_tx);
+
+	return 1;
+}
+
+#define SPI_INGENIC_TX(x)							\
+static int spi_ingenic_tx##x(struct ingenic_spi *priv,				\
+			     struct spi_transfer *xfer)				\
+{										\
+	unsigned int count = xfer->len / (x / 8);				\
+	unsigned int prefill = min(count, SPI_INGENIC_FIFO_SIZE);		\
+	const u##x *tx_buf = xfer->tx_buf;					\
+	u##x *rx_buf = xfer->rx_buf;						\
+	unsigned int i, val;							\
+	int err;								\
+										\
+	/* Fill up the TX fifo */						\
+	for (i = 0; i < prefill; i++) {						\
+		val = tx_buf ? tx_buf[i] : 0;					\
+										\
+		regmap_write(priv->map, REG_SSIDR, val);			\
+	}									\
+										\
+	for (i = 0; i < count; i++) {						\
+		err = spi_ingenic_wait(priv, REG_SSISR_RFE, false);		\
+		if (err)							\
+			return err;						\
+										\
+		regmap_read(priv->map, REG_SSIDR, &val);			\
+		if (rx_buf)							\
+			rx_buf[i] = val;					\
+										\
+		if (i < count - prefill) {					\
+			val = tx_buf ? tx_buf[i + prefill] : 0;			\
+										\
+			regmap_write(priv->map, REG_SSIDR, val);		\
+		}								\
+	}									\
+										\
+	return 0;								\
+}
+SPI_INGENIC_TX(8)
+SPI_INGENIC_TX(16)
+SPI_INGENIC_TX(32)
+#undef SPI_INGENIC_TX
+
+static int spi_ingenic_transfer_one(struct spi_controller *ctlr,
+				    struct spi_device *spi,
+				    struct spi_transfer *xfer)
+{
+	struct ingenic_spi *priv = spi_controller_get_devdata(ctlr);
+	unsigned int bits = xfer->bits_per_word ?: spi->bits_per_word;
+	bool can_dma = ctlr->can_dma && ctlr->can_dma(ctlr, spi, xfer);
+
+	spi_ingenic_prepare_transfer(priv, spi, xfer);
+
+	if (ctlr->cur_msg_mapped && can_dma)
+		return spi_ingenic_dma_tx(ctlr, xfer, bits);
+
+	if (bits > 16)
+		return spi_ingenic_tx32(priv, xfer);
+
+	if (bits > 8)
+		return spi_ingenic_tx16(priv, xfer);
+
+	return spi_ingenic_tx8(priv, xfer);
+}
+
+static int spi_ingenic_prepare_message(struct spi_controller *ctlr,
+				       struct spi_message *message)
+{
+	struct ingenic_spi *priv = spi_controller_get_devdata(ctlr);
+	struct spi_device *spi = message->spi;
+	unsigned int cs = REG_SSICR1_FRMHL << spi->chip_select;
+	unsigned int ssicr0_mask = REG_SSICR0_LOOP | REG_SSICR0_FSEL;
+	unsigned int ssicr1_mask = REG_SSICR1_PHA | REG_SSICR1_POL | cs;
+	unsigned int ssicr0 = 0, ssicr1 = 0;
+
+	if (priv->soc_info->has_trendian) {
+		ssicr0_mask |= REG_SSICR0_RENDIAN_LSB | REG_SSICR0_TENDIAN_LSB;
+
+		if (spi->mode & SPI_LSB_FIRST)
+			ssicr0 |= REG_SSICR0_RENDIAN_LSB | REG_SSICR0_TENDIAN_LSB;
+	} else {
+		ssicr1_mask |= REG_SSICR1_LFST;
+
+		if (spi->mode & SPI_LSB_FIRST)
+			ssicr1 |= REG_SSICR1_LFST;
+	}
+
+	if (spi->mode & SPI_LOOP)
+		ssicr0 |= REG_SSICR0_LOOP;
+	if (spi->chip_select)
+		ssicr0 |= REG_SSICR0_FSEL;
+
+	if (spi->mode & SPI_CPHA)
+		ssicr1 |= REG_SSICR1_PHA;
+	if (spi->mode & SPI_CPOL)
+		ssicr1 |= REG_SSICR1_POL;
+	if (spi->mode & SPI_CS_HIGH)
+		ssicr1 |= cs;
+
+	regmap_update_bits(priv->map, REG_SSICR0, ssicr0_mask, ssicr0);
+	regmap_update_bits(priv->map, REG_SSICR1, ssicr1_mask, ssicr1);
+
+	return 0;
+}
+
+static int spi_ingenic_prepare_hardware(struct spi_controller *ctlr)
+{
+	struct ingenic_spi *priv = spi_controller_get_devdata(ctlr);
+	int ret;
+
+	ret = clk_prepare_enable(priv->clk);
+	if (ret)
+		return ret;
+
+	regmap_write(priv->map, REG_SSICR0, REG_SSICR0_EACLRUN);
+	regmap_write(priv->map, REG_SSICR1, 0);
+	regmap_write(priv->map, REG_SSISR, 0);
+	regmap_set_bits(priv->map, REG_SSICR0, REG_SSICR0_SSIE);
+
+	return 0;
+}
+
+static int spi_ingenic_unprepare_hardware(struct spi_controller *ctlr)
+{
+	struct ingenic_spi *priv = spi_controller_get_devdata(ctlr);
+
+	regmap_clear_bits(priv->map, REG_SSICR0, REG_SSICR0_SSIE);
+
+	clk_disable_unprepare(priv->clk);
+
+	return 0;
+}
+
+static bool spi_ingenic_can_dma(struct spi_controller *ctlr,
+				struct spi_device *spi,
+				struct spi_transfer *xfer)
+{
+	struct dma_slave_caps caps;
+	int ret;
+
+	ret = dma_get_slave_caps(ctlr->dma_tx, &caps);
+	if (ret) {
+		dev_err(&spi->dev, "Unable to get slave caps: %d\n", ret);
+		return false;
+	}
+
+	return !caps.max_sg_burst ||
+		xfer->len <= caps.max_sg_burst * SPI_INGENIC_FIFO_SIZE;
+}
+
+static int spi_ingenic_request_dma(struct spi_controller *ctlr,
+				   struct device *dev)
+{
+	ctlr->dma_tx = dma_request_slave_channel(dev, "tx");
+	if (!ctlr->dma_tx)
+		return -ENODEV;
+
+	ctlr->dma_rx = dma_request_slave_channel(dev, "rx");
+
+	if (!ctlr->dma_rx)
+		return -ENODEV;
+
+	ctlr->can_dma = spi_ingenic_can_dma;
+
+	return 0;
+}
+
+static void spi_ingenic_release_dma(void *data)
+{
+	struct spi_controller *ctlr = data;
+
+	if (ctlr->dma_tx)
+		dma_release_channel(ctlr->dma_tx);
+	if (ctlr->dma_rx)
+		dma_release_channel(ctlr->dma_rx);
+}
+
+static const struct regmap_config spi_ingenic_regmap_config = {
+	.reg_bits = 32,
+	.val_bits = 32,
+	.reg_stride = 4,
+	.max_register = REG_SSIGR,
+};
+
+static int spi_ingenic_probe(struct platform_device *pdev)
+{
+	const struct jz_soc_info *pdata;
+	struct device *dev = &pdev->dev;
+	struct spi_controller *ctlr;
+	struct ingenic_spi *priv;
+	void __iomem *base;
+	int ret;
+
+	pdata = of_device_get_match_data(dev);
+	if (!pdata) {
+		dev_err(dev, "Missing platform data.\n");
+		return -EINVAL;
+	}
+
+	ctlr = devm_spi_alloc_master(dev, sizeof(*priv));
+	if (!ctlr) {
+		dev_err(dev, "Unable to allocate SPI controller.\n");
+		return -ENOMEM;
+	}
+
+	priv = spi_controller_get_devdata(ctlr);
+	priv->soc_info = pdata;
+
+	priv->clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(priv->clk)) {
+		return dev_err_probe(dev, PTR_ERR(priv->clk),
+				     "Unable to get clock.\n");
+	}
+
+	base = devm_platform_get_and_ioremap_resource(pdev, 0, &priv->mem_res);
+	if (IS_ERR(base))
+		return PTR_ERR(base);
+
+	priv->map = devm_regmap_init_mmio(dev, base, &spi_ingenic_regmap_config);
+	if (IS_ERR(priv->map))
+		return PTR_ERR(priv->map);
+
+	priv->flen_field = devm_regmap_field_alloc(dev, priv->map,
+						   pdata->flen_field);
+	if (IS_ERR(priv->flen_field))
+		return PTR_ERR(priv->flen_field);
+
+	platform_set_drvdata(pdev, ctlr);
+
+	ctlr->prepare_transfer_hardware = spi_ingenic_prepare_hardware;
+	ctlr->unprepare_transfer_hardware = spi_ingenic_unprepare_hardware;
+	ctlr->prepare_message = spi_ingenic_prepare_message;
+	ctlr->set_cs = spi_ingenic_set_cs;
+	ctlr->transfer_one = spi_ingenic_transfer_one;
+	ctlr->mode_bits = SPI_MODE_3 | SPI_LSB_FIRST | SPI_LOOP | SPI_CS_HIGH;
+	ctlr->flags = SPI_CONTROLLER_MUST_RX | SPI_CONTROLLER_MUST_TX;
+	ctlr->max_dma_len = SPI_INGENIC_FIFO_SIZE;
+	ctlr->bits_per_word_mask = pdata->bits_per_word_mask;
+	ctlr->min_speed_hz = 7200;
+	ctlr->max_speed_hz = 54000000;
+	ctlr->num_chipselect = 2;
+	ctlr->dev.of_node = pdev->dev.of_node;
+
+	if (spi_ingenic_request_dma(ctlr, dev))
+		dev_warn(dev, "DMA not available.\n");
+
+	ret = devm_add_action_or_reset(dev, spi_ingenic_release_dma, ctlr);
+	if (ret) {
+		dev_err(dev, "Unable to add action.\n");
+		return ret;
+	}
+
+	ret = devm_spi_register_controller(dev, ctlr);
+	if (ret)
+		dev_err(dev, "Unable to register SPI controller.\n");
+
+	return ret;
+}
+
+static const struct jz_soc_info jz4750_soc_info = {
+	.bits_per_word_mask = SPI_BPW_RANGE_MASK(2, 17),
+	.flen_field = REG_FIELD(REG_SSICR1, 4, 7),
+	.has_trendian = false,
+};
+
+static const struct jz_soc_info jz4780_soc_info = {
+	.bits_per_word_mask = SPI_BPW_RANGE_MASK(2, 32),
+	.flen_field = REG_FIELD(REG_SSICR1, 3, 7),
+	.has_trendian = true,
+};
+
+static const struct of_device_id spi_ingenic_of_match[] = {
+	{ .compatible = "ingenic,jz4750-spi", .data = &jz4750_soc_info },
+	{ .compatible = "ingenic,jz4780-spi", .data = &jz4780_soc_info },
+	{}
+};
+MODULE_DEVICE_TABLE(of, spi_ingenic_of_match);
+
+static struct platform_driver spi_ingenic_driver = {
+	.driver = {
+		.name = "spi-ingenic",
+		.of_match_table = spi_ingenic_of_match,
+	},
+	.probe = spi_ingenic_probe,
+};
+
+module_platform_driver(spi_ingenic_driver);
+MODULE_DESCRIPTION("SPI bus driver for the Ingenic JZ47xx SoCs");
+MODULE_AUTHOR("Artur Rojek <contact@artur-rojek.eu>");
+MODULE_AUTHOR("Paul Cercueil <paul@crapouillou.net>");
+MODULE_LICENSE("GPL");
--- a/drivers/spi/spi-mtk-nor.c
+++ b/drivers/spi/spi-mtk-nor.c
@ -160,7 +160,7 @@ static bool mtk_nor_match_read(const struct spi_mem_op *op)
 {
 	int dummy = 0;

-	if (op->dummy.buswidth)
+	if (op->dummy.nbytes)
 		dummy = op->dummy.nbytes * BITS_PER_BYTE / op->dummy.buswidth;

 	if ((op->data.buswidth == 2) || (op->data.buswidth == 4)) {
--- a/drivers/spi/spi-orion.c
+++ b/drivers/spi/spi-orion.c
@ -769,6 +769,7 @@ static int orion_spi_probe(struct platform_device *pdev)
 		dir_acc->vaddr = devm_ioremap(&pdev->dev, r->start, PAGE_SIZE);
 		if (!dir_acc->vaddr) {
 			status = -ENOMEM;
+			of_node_put(np);
 			goto out_rel_axi_clk;
 		}
 		dir_acc->size = PAGE_SIZE;
--- a/drivers/spi/spi-rpc-if.c
+++ b/drivers/spi/spi-rpc-if.c
@ -139,7 +139,9 @@ static int rpcif_spi_probe(struct platform_device *pdev)
 		return -ENOMEM;

 	rpc = spi_controller_get_devdata(ctlr);
-	rpcif_sw_init(rpc, parent);
+	error = rpcif_sw_init(rpc, parent);
+	if (error)
+		return error;

 	platform_set_drvdata(pdev, ctlr);

--- a/drivers/spi/spi-rspi.c
+++ b/drivers/spi/spi-rspi.c
@ -1427,4 +1427,3 @@ module_platform_driver(rspi_driver);
 MODULE_DESCRIPTION("Renesas RSPI bus driver");
 MODULE_LICENSE("GPL v2");
 MODULE_AUTHOR("Yoshihiro Shimoda");
-MODULE_ALIAS("platform:rspi");
--- a/drivers/spi/spi-sh-msiof.c
+++ b/drivers/spi/spi-sh-msiof.c
@ -1426,4 +1426,3 @@ module_platform_driver(sh_msiof_spi_drv);
 MODULE_DESCRIPTION("SuperH MSIOF SPI Controller Interface Driver");
 MODULE_AUTHOR("Magnus Damm");
 MODULE_LICENSE("GPL v2");
-MODULE_ALIAS("platform:spi_sh_msiof");
--- a/drivers/spi/spi-stm32-qspi.c
+++ b/drivers/spi/spi-stm32-qspi.c
@ -397,7 +397,7 @@ static int stm32_qspi_send(struct spi_mem *mem, const struct spi_mem_op *op)
 		ccr |= FIELD_PREP(CCR_ADSIZE_MASK, op->addr.nbytes - 1);
 	}

-	if (op->dummy.buswidth && op->dummy.nbytes)
+	if (op->dummy.nbytes)
 		ccr |= FIELD_PREP(CCR_DCYC_MASK,
 				  op->dummy.nbytes * 8 / op->dummy.buswidth);

--- a/drivers/spi/spi-tegra20-slink.c
+++ b/drivers/spi/spi-tegra20-slink.c
@ -1124,7 +1124,7 @@ exit_free_irq:
 exit_pm_put:
 	pm_runtime_put(&pdev->dev);
 exit_pm_disable:
-	pm_runtime_disable(&pdev->dev);
+	pm_runtime_force_suspend(&pdev->dev);

 	tegra_slink_deinit_dma_param(tspi, false);
 exit_rx_dma_free:
@ -1143,7 +1143,7 @@ static int tegra_slink_remove(struct platform_device *pdev)

 	free_irq(tspi->irq, tspi);

-	pm_runtime_disable(&pdev->dev);
+	pm_runtime_force_suspend(&pdev->dev);

 	if (tspi->tx_dma_chan)
 		tegra_slink_deinit_dma_param(tspi, false);
--- a/drivers/spi/spi-tegra210-quad.c
+++ b/drivers/spi/spi-tegra210-quad.c
@ -1318,7 +1318,7 @@ static int tegra_qspi_probe(struct platform_device *pdev)
 exit_free_irq:
 	free_irq(qspi_irq, tqspi);
 exit_pm_disable:
-	pm_runtime_disable(&pdev->dev);
+	pm_runtime_force_suspend(&pdev->dev);
 	tegra_qspi_deinit_dma(tqspi);
 	return ret;
 }
@ -1330,7 +1330,7 @@ static int tegra_qspi_remove(struct platform_device *pdev)

 	spi_unregister_master(master);
 	free_irq(tqspi->irq, tqspi);
-	pm_runtime_disable(&pdev->dev);
+	pm_runtime_force_suspend(&pdev->dev);
 	tegra_qspi_deinit_dma(tqspi);

 	return 0;
--- a/drivers/spi/spi-tle62x0.c
+++ b/drivers/spi/spi-tle62x0.c
@ -141,7 +141,7 @@ static ssize_t tle62x0_gpio_show(struct device *dev,
 	value = (st->gpio_state >> gpio_num) & 1;
 	mutex_unlock(&st->lock);

-	return snprintf(buf, PAGE_SIZE, "%d", value);
+	return sysfs_emit(buf, "%d", value);
 }

 static ssize_t tle62x0_gpio_store(struct device *dev,
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@ -285,7 +285,7 @@ static const struct attribute_group *spi_master_groups[] = {
 	NULL,
 };

-void spi_statistics_add_transfer_stats(struct spi_statistics *stats,
+static void spi_statistics_add_transfer_stats(struct spi_statistics *stats,
 					      struct spi_transfer *xfer,
 					      struct spi_controller *ctlr)
 {
@ -310,7 +310,6 @@ void spi_statistics_add_transfer_stats(struct spi_statistics *stats,

 	spin_unlock_irqrestore(&stats->lock, flags);
 }
-EXPORT_SYMBOL_GPL(spi_statistics_add_transfer_stats);

 /* modalias support makes "modprobe $MODALIAS" new-style hotplug work,
 * and the sysfs version makes coldplug work too.
@ -536,7 +535,7 @@ static DEFINE_MUTEX(board_lock);
 *
 * Return: a pointer to the new device, or NULL.
 */
-struct spi_device *spi_alloc_device(struct spi_controller *ctlr)
+static struct spi_device *spi_alloc_device(struct spi_controller *ctlr)
 {
 	struct spi_device	*spi;

@ -561,7 +560,6 @@ struct spi_device *spi_alloc_device(struct spi_controller *ctlr)
 	device_initialize(&spi->dev);
 	return spi;
 }
-EXPORT_SYMBOL_GPL(spi_alloc_device);

 static void spi_dev_set_name(struct spi_device *spi)
 {
@ -599,6 +597,11 @@ static int __spi_add_device(struct spi_device *spi)
 	struct device *dev = ctlr->dev.parent;
 	int status;

+	/*
+	 * We need to make sure there's no other device with this
+	 * chipselect **BEFORE** we call setup(), else we'll trash
+	 * its configuration.
+	 */
 	status = bus_for_each_dev(&spi_bus_type, NULL, spi, spi_dev_check);
 	if (status) {
 		dev_err(dev, "chipselect %d already in use\n",
@ -651,7 +654,7 @@ static int __spi_add_device(struct spi_device *spi)
 *
 * Return: 0 on success; negative errno on failure
 */
-int spi_add_device(struct spi_device *spi)
+static int spi_add_device(struct spi_device *spi)
 {
 	struct spi_controller *ctlr = spi->controller;
 	struct device *dev = ctlr->dev.parent;
@ -667,16 +670,11 @@ int spi_add_device(struct spi_device *spi)
 	/* Set the bus ID string */
 	spi_dev_set_name(spi);

-	/* We need to make sure there's no other device with this
-	 * chipselect **BEFORE** we call setup(), else we'll trash
-	 * its configuration.  Lock against concurrent add() calls.
-	 */
 	mutex_lock(&ctlr->add_lock);
 	status = __spi_add_device(spi);
 	mutex_unlock(&ctlr->add_lock);
 	return status;
 }
-EXPORT_SYMBOL_GPL(spi_add_device);

 static int spi_add_device_locked(struct spi_device *spi)
 {
@ -851,6 +849,87 @@ int spi_register_board_info(struct spi_board_info const *info, unsigned n)

 /*-------------------------------------------------------------------------*/

+/* Core methods for SPI resource management */
+
+/**
+ * spi_res_alloc - allocate a spi resource that is life-cycle managed
+ *                 during the processing of a spi_message while using
+ *                 spi_transfer_one
+ * @spi:     the spi device for which we allocate memory
+ * @release: the release code to execute for this resource
+ * @size:    size to alloc and return
+ * @gfp:     GFP allocation flags
+ *
+ * Return: the pointer to the allocated data
+ *
+ * This may get enhanced in the future to allocate from a memory pool
+ * of the @spi_device or @spi_controller to avoid repeated allocations.
+ */
+static void *spi_res_alloc(struct spi_device *spi, spi_res_release_t release,
+			   size_t size, gfp_t gfp)
+{
+	struct spi_res *sres;
+
+	sres = kzalloc(sizeof(*sres) + size, gfp);
+	if (!sres)
+		return NULL;
+
+	INIT_LIST_HEAD(&sres->entry);
+	sres->release = release;
+
+	return sres->data;
+}
+
+/**
+ * spi_res_free - free an spi resource
+ * @res: pointer to the custom data of a resource
+ *
+ */
+static void spi_res_free(void *res)
+{
+	struct spi_res *sres = container_of(res, struct spi_res, data);
+
+	if (!res)
+		return;
+
+	WARN_ON(!list_empty(&sres->entry));
+	kfree(sres);
+}
+
+/**
+ * spi_res_add - add a spi_res to the spi_message
+ * @message: the spi message
+ * @res:     the spi_resource
+ */
+static void spi_res_add(struct spi_message *message, void *res)
+{
+	struct spi_res *sres = container_of(res, struct spi_res, data);
+
+	WARN_ON(!list_empty(&sres->entry));
+	list_add_tail(&sres->entry, &message->resources);
+}
+
+/**
+ * spi_res_release - release all spi resources for this message
+ * @ctlr:  the @spi_controller
+ * @message: the @spi_message
+ */
+static void spi_res_release(struct spi_controller *ctlr, struct spi_message *message)
+{
+	struct spi_res *res, *tmp;
+
+	list_for_each_entry_safe_reverse(res, tmp, &message->resources, entry) {
+		if (res->release)
+			res->release(ctlr, message, res->data);
+
+		list_del(&res->entry);
+
+		kfree(res);
+	}
+}
+
+/*-------------------------------------------------------------------------*/
+
 static void spi_set_cs(struct spi_device *spi, bool enable, bool force)
 {
 	bool activate = enable;
@ -3068,127 +3147,6 @@ int spi_controller_resume(struct spi_controller *ctlr)
 }
 EXPORT_SYMBOL_GPL(spi_controller_resume);

-static int __spi_controller_match(struct device *dev, const void *data)
-{
-	struct spi_controller *ctlr;
-	const u16 *bus_num = data;
-
-	ctlr = container_of(dev, struct spi_controller, dev);
-	return ctlr->bus_num == *bus_num;
-}
-
-/**
- * spi_busnum_to_master - look up master associated with bus_num
- * @bus_num: the master's bus number
- * Context: can sleep
- *
- * This call may be used with devices that are registered after
- * arch init time.  It returns a refcounted pointer to the relevant
- * spi_controller (which the caller must release), or NULL if there is
- * no such master registered.
- *
- * Return: the SPI master structure on success, else NULL.
- */
-struct spi_controller *spi_busnum_to_master(u16 bus_num)
-{
-	struct device		*dev;
-	struct spi_controller	*ctlr = NULL;
-
-	dev = class_find_device(&spi_master_class, NULL, &bus_num,
-				__spi_controller_match);
-	if (dev)
-		ctlr = container_of(dev, struct spi_controller, dev);
-	/* reference got in class_find_device */
-	return ctlr;
-}
-EXPORT_SYMBOL_GPL(spi_busnum_to_master);
-
-/*-------------------------------------------------------------------------*/
-
-/* Core methods for SPI resource management */
-
-/**
- * spi_res_alloc - allocate a spi resource that is life-cycle managed
- *                 during the processing of a spi_message while using
- *                 spi_transfer_one
- * @spi:     the spi device for which we allocate memory
- * @release: the release code to execute for this resource
- * @size:    size to alloc and return
- * @gfp:     GFP allocation flags
- *
- * Return: the pointer to the allocated data
- *
- * This may get enhanced in the future to allocate from a memory pool
- * of the @spi_device or @spi_controller to avoid repeated allocations.
- */
-void *spi_res_alloc(struct spi_device *spi,
-		    spi_res_release_t release,
-		    size_t size, gfp_t gfp)
-{
-	struct spi_res *sres;
-
-	sres = kzalloc(sizeof(*sres) + size, gfp);
-	if (!sres)
-		return NULL;
-
-	INIT_LIST_HEAD(&sres->entry);
-	sres->release = release;
-
-	return sres->data;
-}
-EXPORT_SYMBOL_GPL(spi_res_alloc);
-
-/**
- * spi_res_free - free an spi resource
- * @res: pointer to the custom data of a resource
- *
- */
-void spi_res_free(void *res)
-{
-	struct spi_res *sres = container_of(res, struct spi_res, data);
-
-	if (!res)
-		return;
-
-	WARN_ON(!list_empty(&sres->entry));
-	kfree(sres);
-}
-EXPORT_SYMBOL_GPL(spi_res_free);
-
-/**
- * spi_res_add - add a spi_res to the spi_message
- * @message: the spi message
- * @res:     the spi_resource
- */
-void spi_res_add(struct spi_message *message, void *res)
-{
-	struct spi_res *sres = container_of(res, struct spi_res, data);
-
-	WARN_ON(!list_empty(&sres->entry));
-	list_add_tail(&sres->entry, &message->resources);
-}
-EXPORT_SYMBOL_GPL(spi_res_add);
-
-/**
- * spi_res_release - release all spi resources for this message
- * @ctlr:  the @spi_controller
- * @message: the @spi_message
- */
-void spi_res_release(struct spi_controller *ctlr, struct spi_message *message)
-{
-	struct spi_res *res, *tmp;
-
-	list_for_each_entry_safe_reverse(res, tmp, &message->resources, entry) {
-		if (res->release)
-			res->release(ctlr, message, res->data);
-
-		list_del(&res->entry);
-
-		kfree(res);
-	}
-}
-EXPORT_SYMBOL_GPL(spi_res_release);
-
 /*-------------------------------------------------------------------------*/

 /* Core methods for spi_message alterations */
@ -3227,7 +3185,7 @@ static void __spi_replace_transfers_release(struct spi_controller *ctlr,
 * Returns: pointer to @spi_replaced_transfers,
 *          PTR_ERR(...) in case of errors.
 */
-struct spi_replaced_transfers *spi_replace_transfers(
+static struct spi_replaced_transfers *spi_replace_transfers(
 	struct spi_message *msg,
 	struct spi_transfer *xfer_first,
 	size_t remove,
@ -3319,7 +3277,6 @@ struct spi_replaced_transfers *spi_replace_transfers(

 	return rxfer;
 }
-EXPORT_SYMBOL_GPL(spi_replace_transfers);

 static int __spi_split_transfer_maxsize(struct spi_controller *ctlr,
 					struct spi_message *msg,
@ -3869,7 +3826,7 @@ EXPORT_SYMBOL_GPL(spi_async);
 *
 * Return: zero on success, else a negative error code.
 */
-int spi_async_locked(struct spi_device *spi, struct spi_message *message)
+static int spi_async_locked(struct spi_device *spi, struct spi_message *message)
 {
 	struct spi_controller *ctlr = spi->controller;
 	int ret;
@ -3888,7 +3845,6 @@ int spi_async_locked(struct spi_device *spi, struct spi_message *message)
 	return ret;

 }
-EXPORT_SYMBOL_GPL(spi_async_locked);

 /*-------------------------------------------------------------------------*/

@ -4146,18 +4102,15 @@ EXPORT_SYMBOL_GPL(spi_write_then_read);

 /*-------------------------------------------------------------------------*/

-#if IS_ENABLED(CONFIG_OF)
+#if IS_ENABLED(CONFIG_OF_DYNAMIC)
 /* must call put_device() when done with returned spi_device device */
-struct spi_device *of_find_spi_device_by_node(struct device_node *node)
+static struct spi_device *of_find_spi_device_by_node(struct device_node *node)
 {
 	struct device *dev = bus_find_device_by_of_node(&spi_bus_type, node);

 	return dev ? to_spi_device(dev) : NULL;
 }
-EXPORT_SYMBOL_GPL(of_find_spi_device_by_node);
-#endif /* IS_ENABLED(CONFIG_OF) */

-#if IS_ENABLED(CONFIG_OF_DYNAMIC)
 /* the spi controllers are not using spi_bus, so we find it with another way */
 static struct spi_controller *of_find_spi_controller_by_node(struct device_node *node)
 {
--- a/include/linux/firmware/xlnx-zynqmp.h
+++ b/include/linux/firmware/xlnx-zynqmp.h
@ -123,6 +123,7 @@ enum pm_ioctl_id {
 	IOCTL_READ_PGGS = 15,
 	/* Set healthy bit value */
 	IOCTL_SET_BOOT_HEALTH_STATUS = 17,
+	IOCTL_OSPI_MUX_SELECT = 21,
 };

 enum pm_query_id {
@ -351,6 +352,11 @@ enum zynqmp_pm_shutdown_subtype {
 	ZYNQMP_PM_SHUTDOWN_SUBTYPE_SYSTEM = 2,
 };

+enum ospi_mux_select_type {
+	PM_OSPI_MUX_SEL_DMA = 0,
+	PM_OSPI_MUX_SEL_LINEAR = 1,
+};
+
 /**
 * struct zynqmp_pm_query_data - PM query data
 * @qid:	query ID
@ -387,6 +393,7 @@ int zynqmp_pm_set_pll_frac_data(u32 clk_id, u32 data);
 int zynqmp_pm_get_pll_frac_data(u32 clk_id, u32 *data);
 int zynqmp_pm_set_sd_tapdelay(u32 node_id, u32 type, u32 value);
 int zynqmp_pm_sd_dll_reset(u32 node_id, u32 type);
+int zynqmp_pm_ospi_mux_select(u32 dev_id, u32 select);
 int zynqmp_pm_reset_assert(const enum zynqmp_pm_reset reset,
 			   const enum zynqmp_pm_reset_action assert_flag);
 int zynqmp_pm_reset_get_status(const enum zynqmp_pm_reset reset, u32 *status);
@ -508,6 +515,11 @@ static inline int zynqmp_pm_sd_dll_reset(u32 node_id, u32 type)
 	return -ENODEV;
 }

+static inline int zynqmp_pm_ospi_mux_select(u32 dev_id, u32 select)
+{
+	return -ENODEV;
+}
+
 static inline int zynqmp_pm_reset_assert(const enum zynqmp_pm_reset reset,
 					 const enum zynqmp_pm_reset_action assert_flag)
 {
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@ -78,10 +78,6 @@ struct spi_statistics {
 	unsigned long transfers_split_maxsize;
 };

-void spi_statistics_add_transfer_stats(struct spi_statistics *stats,
-				       struct spi_transfer *xfer,
-				       struct spi_controller *ctlr);
-
 #define SPI_STATISTICS_ADD_TO_FIELD(stats, field, count)	\
 	do {							\
 		unsigned long flags;				\
@ -763,8 +759,6 @@ extern int devm_spi_register_controller(struct device *dev,
 					struct spi_controller *ctlr);
 extern void spi_unregister_controller(struct spi_controller *ctlr);

-extern struct spi_controller *spi_busnum_to_master(u16 busnum);
-
 /*
 * SPI resource management while processing a SPI message
 */
@ -788,15 +782,6 @@ struct spi_res {
 	unsigned long long      data[]; /* guarantee ull alignment */
 };

-extern void *spi_res_alloc(struct spi_device *spi,
-			   spi_res_release_t release,
-			   size_t size, gfp_t gfp);
-extern void spi_res_add(struct spi_message *message, void *res);
-extern void spi_res_free(void *res);
-
-extern void spi_res_release(struct spi_controller *ctlr,
-			    struct spi_message *message);
-
 /*---------------------------------------------------------------------------*/

 /*
@ -1114,8 +1099,6 @@ static inline void spi_message_free(struct spi_message *m)

 extern int spi_setup(struct spi_device *spi);
 extern int spi_async(struct spi_device *spi, struct spi_message *message);
-extern int spi_async_locked(struct spi_device *spi,
-			    struct spi_message *message);
 extern int spi_slave_abort(struct spi_device *spi);

 static inline size_t
@ -1198,15 +1181,6 @@ struct spi_replaced_transfers {
 	struct spi_transfer inserted_transfers[];
 };

-extern struct spi_replaced_transfers *spi_replace_transfers(
-	struct spi_message *msg,
-	struct spi_transfer *xfer_first,
-	size_t remove,
-	size_t insert,
-	spi_replaced_release_t release,
-	size_t extradatasize,
-	gfp_t gfp);
-
 /*---------------------------------------------------------------------------*/

 /* SPI transfer transformation methods */
@ -1478,19 +1452,7 @@ spi_register_board_info(struct spi_board_info const *info, unsigned n)
 * use spi_new_device() to describe each device.  You can also call
 * spi_unregister_device() to start making that device vanish, but
 * normally that would be handled by spi_unregister_controller().
- *
- * You can also use spi_alloc_device() and spi_add_device() to use a two
- * stage registration sequence for each spi_device.  This gives the caller
- * some more control over the spi_device structure before it is registered,
- * but requires that caller to initialize fields that would otherwise
- * be defined using the board info.
 */
-extern struct spi_device *
-spi_alloc_device(struct spi_controller *ctlr);
-
-extern int
-spi_add_device(struct spi_device *spi);
-
 extern struct spi_device *
 spi_new_device(struct spi_controller *, struct spi_board_info *);

@ -1505,23 +1467,6 @@ spi_transfer_is_last(struct spi_controller *ctlr, struct spi_transfer *xfer)
 	return list_is_last(&xfer->transfer_list, &ctlr->cur_msg->transfers);
 }

-/* OF support code */
-#if IS_ENABLED(CONFIG_OF)
-
-/* must call put_device() when done with returned spi_device device */
-extern struct spi_device *
-of_find_spi_device_by_node(struct device_node *node);
-
-#else
-
-static inline struct spi_device *
-of_find_spi_device_by_node(struct device_node *node)
-{
-	return NULL;
-}
-
-#endif /* IS_ENABLED(CONFIG_OF) */
-
 /* Compatibility layer */
 #define spi_master			spi_controller