[PATCH v5 33/33] riscv: Add Sipeed Maix support

[Sean Anderson <seanga2@gmail.com>]

The Sipeed Maix series is a collection of boards built around the RISC-V
Kendryte K210 processor. This processor contains several peripherals to
accelerate neural network processing and other "ai" tasks. This includes a
"KPU" neural network processor, an audio processor supporting beamforming
reception, and a digital video port supporting capture and output at VGA
resolution. Other peripherals include 8M of sram (accessible with and
without caching); remappable pins, including 40 GPIOs; AES, FFT, and SHA256
accelerators; a DMA controller; and I2C, I2S, and SPI controllers. Maix
peripherals vary, but include spi flash; on-board usb-serial bridges; ports
for cameras, displays, and sd cards; and ESP32 chips. Currently, only the
Sipeed Maix Bit V2.0 (bitm) is supported, but the boards are fairly
similar.

Documentation for Maix boards is located at
<http://dl.sipeed.com/MAIX/HDK/>.  Documentation for the Kendryte K210 is
located at <https://kendryte.com/downloads/>. However, hardware details are
rather lacking, so most technical reference has been taken from the
standalone sdk located at
<https://github.com/kendryte/kendryte-standalone-sdk>.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
---

Changes in v5:
- Configure relocation location with CONFIG_SYS_SDRAM_*
- Enable ram clocks
- Add pinmux/gpio/led support
- Remove (broken) MMC support
- Store the environment in flash
- Add partitions
- Add bootcmd
- Add docs for pinctrl and booting

Changes in v4:
- Rework documentation to be organized by board mfg not cpu mfg
- Update docs to reflect working SPI support
- Add proper spi support
- Don't define unneecessary macros in config.h
- Lower the default stack so it isn't clobbered on relocation
- Update MAINTAINERS
- Update copyright

Changes in v3:
- Reorder to be last in the patch series
- Add documentation for the board
- Generate defconfig with "make savedefconfig"
- Update Kconfig to imply most features we need
- Update MAINTAINERS

Changes in v2:
- Select CONFIG_SYS_RISCV_NOCOUNTER
- Imply CONFIG_CLK_K210
- Remove spurious references to CONFIG_ARCH_K210
- Remove many configs from defconfig where the defaults were fine
- Add a few "not set" lines to suppress unneeded defaults
- Reduce pre-reloc malloc space, now that clocks initialization happens
  later

 arch/riscv/Kconfig                 |   4 +
 board/sipeed/maix/Kconfig          |  72 ++++++++++
 board/sipeed/maix/MAINTAINERS      |  11 ++
 board/sipeed/maix/Makefile         |   5 +
 board/sipeed/maix/maix.c           |  54 +++++++
 configs/sipeed_maix_bitm_defconfig |  16 +++
 doc/board/index.rst                |   1 +
 doc/board/sipeed/index.rst         |   9 ++
 doc/board/sipeed/maix.rst          | 223 +++++++++++++++++++++++++++++
 include/configs/sipeed-maix.h      |  24 ++++
 10 files changed, 419 insertions(+)
 create mode 100644 board/sipeed/maix/Kconfig
 create mode 100644 board/sipeed/maix/MAINTAINERS
 create mode 100644 board/sipeed/maix/Makefile
 create mode 100644 board/sipeed/maix/maix.c
 create mode 100644 configs/sipeed_maix_bitm_defconfig
 create mode 100644 doc/board/sipeed/index.rst
 create mode 100644 doc/board/sipeed/maix.rst
 create mode 100644 include/configs/sipeed-maix.h

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index b7a5757584..d016dd75d7 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -20,6 +20,9 @@ config TARGET_QEMU_VIRT
 config TARGET_SIFIVE_FU540
 	bool "Support SiFive FU540 Board"
 
+config TARGET_SIPEED_MAIX
+	bool "Support Sipeed Maix Board"
+
 endchoice
 
 config SYS_ICACHE_OFF
@@ -53,6 +56,7 @@ source "board/AndesTech/ax25-ae350/Kconfig"
 source "board/emulation/qemu-riscv/Kconfig"
 source "board/microchip/mpfs_icicle/Kconfig"
 source "board/sifive/fu540/Kconfig"
+source "board/sipeed/maix/Kconfig"
 
 # platform-specific options below
 source "arch/riscv/cpu/ax25/Kconfig"
diff --git a/board/sipeed/maix/Kconfig b/board/sipeed/maix/Kconfig
new file mode 100644
index 0000000000..939eb4829a
--- /dev/null
+++ b/board/sipeed/maix/Kconfig
@@ -0,0 +1,72 @@
+# SPDX-License-Identifier: GPL-2.0+
+# Copyright (C) 2019-20 Sean Anderson <seanga2@gmail.com>
+
+if TARGET_SIPEED_MAIX
+
+config SYS_BOARD
+	default "maix"
+
+config SYS_VENDOR
+	default "sipeed"
+
+config SYS_CPU
+	default "generic"
+
+config SYS_CONFIG_NAME
+	default "sipeed-maix"
+
+config SYS_TEXT_BASE
+	default 0x80000000
+
+config DEFAULT_DEVICE_TREE
+	default "k210-maix-bit"
+
+config NR_CPUS
+	default 2
+
+config NR_DRAM_BANKS
+	default 3
+
+config SF_DEFAULT_BUS
+	default 3
+
+config BOARD_SPECIFIC_OPTIONS
+	def_bool y
+	select GENERIC_RISCV
+	select RISCV_PRIV_1_9
+	imply SMP
+	imply OF_BOARD_SETUP
+	imply DM_SERIAL
+	imply SIFIVE_SERIAL
+	imply SIFIVE_CLINT
+	imply POWER_DOMAIN
+	imply SIMPLE_PM_BUS
+	imply CLK_CCF
+	imply CLK_COMPOSITE_CCF
+	imply CLK_K210
+	imply DM_RESET
+	imply RESET_SYSCON
+	imply SYSRESET
+	imply SYSRESET_SYSCON
+	imply PINCTRL
+	imply PINCONF
+	imply PINCTRL_K210
+	imply DM_GPIO
+	imply DWAPB_GPIO
+	imply SIFIVE_GPIO
+	imply CMD_GPIO
+	imply LED
+	imply LED_GPIO
+	imply SPI
+	imply DESIGNWARE_SPI
+	imply SPI_FLASH_WINBOND
+	imply DM_MTD
+	imply SPI_FLASH_MTD
+	imply CMD_MTD
+	imply ENV_IS_IN_SPI_FLASH
+	imply MMC
+	imply MMC_BROKEN_CD
+	imply MMC_SPI
+	imply WDT
+	imply DESIGNWARE_WATCHDOG
+endif
diff --git a/board/sipeed/maix/MAINTAINERS b/board/sipeed/maix/MAINTAINERS
new file mode 100644
index 0000000000..1f33882e1e
--- /dev/null
+++ b/board/sipeed/maix/MAINTAINERS
@@ -0,0 +1,11 @@
+Sipeed Maix BOARD
+M:	Sean Anderson <seanga2@gmail.com>
+S:	Maintained
+F:	arch/riscv/dts/k210.dtsi
+F:	arch/riscv/dts/k210-maix-bit.dts
+F:	board/sipeed/maix/
+F:	configs/sipeed_maix_defconfig
+F:	doc/board/sipeed/
+F:	include/configs/sipeed-maix.h
+F:	include/dt-bindings/*/k210-sysctl.h
+F:	test/dm/k210_pll.c
diff --git a/board/sipeed/maix/Makefile b/board/sipeed/maix/Makefile
new file mode 100644
index 0000000000..4acff5b31e
--- /dev/null
+++ b/board/sipeed/maix/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0+
+#
+# Copyright (c) 2019 Western Digital Corporation or its affiliates.
+
+obj-y += maix.o
diff --git a/board/sipeed/maix/maix.c b/board/sipeed/maix/maix.c
new file mode 100644
index 0000000000..c126cb5d67
--- /dev/null
+++ b/board/sipeed/maix/maix.c
@@ -0,0 +1,54 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright (C) 2019 Sean Anderson <seanga2@gmail.com>
+ */
+
+#include <common.h>
+#include <clk.h>
+#include <dm.h>
+#include <fdt_support.h>
+
+phys_size_t get_effective_memsize(void)
+{
+	return CONFIG_SYS_SDRAM_SIZE;
+}
+
+int board_init(void)
+{
+	int ret;
+	ofnode bank = ofnode_null();
+
+	/* Enable RAM clocks */
+	while (true) {
+		struct clk clk;
+
+		bank = ofnode_by_prop_value(bank, "device_type", "memory",
+					    sizeof("memory"));
+		if (ofnode_equal(bank, ofnode_null()))
+			break;
+
+		ret = clk_get_by_index_nodev(bank, 0, &clk);
+		if (ret)
+			continue;
+
+		ret = clk_enable(&clk);
+		clk_free(&clk);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+int ft_board_setup(void *blob, bd_t *bd)
+{
+	int i;
+	u64 base[CONFIG_NR_DRAM_BANKS];
+	u64 size[CONFIG_NR_DRAM_BANKS];
+
+	for (i = 0; i < CONFIG_NR_DRAM_BANKS; i++) {
+		base[i] = bd->bi_dram[i].start;
+		size[i] = bd->bi_dram[i].size;
+	}
+
+	return fdt_fixup_memory_banks(blob, base, size, CONFIG_NR_DRAM_BANKS);
+}
diff --git a/configs/sipeed_maix_bitm_defconfig b/configs/sipeed_maix_bitm_defconfig
new file mode 100644
index 0000000000..7f644e7a37
--- /dev/null
+++ b/configs/sipeed_maix_bitm_defconfig
@@ -0,0 +1,16 @@
+CONFIG_RISCV=y
+CONFIG_ENV_SIZE=0x2000
+CONFIG_ENV_SECT_SIZE=0x1000
+CONFIG_ENV_OFFSET=0x7C000
+CONFIG_ENV_OFFSET_REDUND=0x7E000
+CONFIG_TARGET_SIPEED_MAIX=y
+CONFIG_ARCH_RV64I=y
+CONFIG_USE_BOOTCOMMAND=y
+CONFIG_BOOTCOMMAND="sf probe;mtd read kernel 80000000;go 80000000"
+CONFIG_MTDIDS_DEFAULT="nor0=spi3.0"
+CONFIG_MTDPARTS_DEFAULT="spi3.0:496k(u-boot),16k(env),5632k(kernel),10240k(data)"
+CONFIG_SYS_REDUNDAND_ENVIRONMENT=y
+# CONFIG_NET is not set
+# CONFIG_INPUT is not set
+# CONFIG_DM_ETH is not set
+# CONFIG_EFI_LOADER is not set
diff --git a/doc/board/index.rst b/doc/board/index.rst
index b8b956d730..54cbc5c874 100644
--- a/doc/board/index.rst
+++ b/doc/board/index.rst
@@ -16,4 +16,5 @@ Board-specific doc
    renesas/index
    rockchip/index
    sifive/index
+   sipeed/index
    xilinx/index
diff --git a/doc/board/sipeed/index.rst b/doc/board/sipeed/index.rst
new file mode 100644
index 0000000000..92da47fddd
--- /dev/null
+++ b/doc/board/sipeed/index.rst
@@ -0,0 +1,9 @@
+.. SPDX-License-Identifier: GPL-2.0+
+
+Kendryte
+========
+
+.. toctree::
+   :maxdepth: 2
+
+   maix
diff --git a/doc/board/sipeed/maix.rst b/doc/board/sipeed/maix.rst
new file mode 100644
index 0000000000..c5c2e24ea5
--- /dev/null
+++ b/doc/board/sipeed/maix.rst
@@ -0,0 +1,223 @@
+.. SPDX-License-Identifier: GPL-2.0+
+.. Copyright (C) 2020 Sean Anderson <seanga2@gmail.com>
+
+Maix Bit
+========
+
+Several of the Sipeed Maix series of boards cotain the Kendryte K210 processor,
+a 64-bit RISC-V CPU. This processor contains several peripherals to accelerate
+neural network processing and other "ai" tasks. This includes a "KPU" neural
+network processor, an audio processor supporting beamforming reception, and a
+digital video port supporting capture and output at VGA resolution. Other
+peripherals include 8M of SRAM (accessible with and without caching); remappable
+pins, including 40 GPIOs; AES, FFT, and SHA256 accelerators; a DMA controller;
+and I2C, I2S, and SPI controllers. Maix peripherals vary, but include spi flash;
+on-board usb-serial bridges; ports for cameras, displays, and sd cards; and
+ESP32 chips. Currently, only the Sipeed Maix Bit V2.0 (bitm) is supported, but
+the boards are fairly similar.
+
+Documentation for Maix boards is available from
+`Sipeed's website <http://dl.sipeed.com/MAIX/HDK/>`_.
+Documentation for the Kendryte K210 is available from
+`Kendryte's website <https://kendryte.com/downloads/>`_. However, hardware
+details are rather lacking, so most technical reference has been taken from the
+`standalone sdk <https://github.com/kendryte/kendryte-standalone-sdk>`_.
+
+Build and boot steps
+--------------------
+
+To build u-boot, run
+
+.. code-block:: none
+
+    make sipeed_maix_bitm_defconfig
+    make CROSS_COMPILE=<your cross compile prefix>
+
+To flash u-boot to a maix bit, run
+
+.. code-block:: none
+
+    kflash -tp /dev/<your tty here> -B bit_mic u-boot-dtb.bin
+
+Boot output should look like the following:
+
+.. code-block:: none
+
+    
+    U-Boot 2020.04-rc2-00087-g2221cc09c1-dirty (Feb 28 2020 - 13:53:09 -0500)
+    
+    DRAM:  8 MiB
+    WDT:   Started with servicing (60s timeout)
+    MMC:   spi at 53000000:slot at 0: 0
+    In:    serial at 38000000
+    Out:   serial at 38000000
+    Err:   serial at 38000000
+    Hit any key to stop autoboot:  0
+    SF: Detected w25q128fw with page size 256 Bytes, erase size 4 KiB, total 16 MiB
+    Reading 5242880 byte(s) at offset 0x00000000
+    ## Starting application at 0x80000000 ...
+
+Flashing Images
+---------------
+
+To flash a kernel, transfer it over serial, then write it to the kernel
+partition.
+
+.. code-block:: none
+
+    => loady 80000000 1500000
+    ## Switch baudrate to 1500000 bps and press ENTER ...
+    
+    *** baud: 1500000
+    
+    *** baud: 1500000 ***
+    ## Ready for binary (ymodem) download to 0x80000000 at 1500000 bps...
+    C
+    *** file: loader.bin
+    $ sz -vv loader.bin
+    Sending: loader.bin
+    Bytes Sent:2478208   BPS:72937                           
+    Sending: 
+    Ymodem sectors/kbytes sent:   0/ 0k
+    Transfer complete
+    
+    *** exit status: 0 ***
+    ## Total Size      = 0x0025d052 = 2478162 Bytes
+    ## Switch baudrate to 115200 bps and press ESC ...
+    
+    *** baud: 115200
+    
+    *** baud: 115200 ***
+    => sf probe
+    SF: Detected w25q128fw with page size 256 Bytes, erase size 4 KiB, total 16 MiB
+    => mtd write kernel 80000000 0 25d052
+    Writing 2478162 byte(s) at offset 0x00000000
+
+**NB:** kflash adds a 5-byte header to payloads (and a 32-byte trailer) to all
+payloads it flashes. If you use kflash to flash your payload, you will need to
+account for this header when specifying what offset in spi flash to load from.
+
+Partition Scheme
+^^^^^^^^^^^^^^^^
+
+There is no partition scheme specified by the manufacturer. The only requirement
+imposed by the firmware is that offset 0 will be loaded and ran. The default
+partition scheme is
+
+========= ======== ======
+Partition Offset   Size
+========= ======== ======
+u-boot    0x000000 496k
+env       0x07C000 16k
+kernel    0x080000 5M
+data      0x580000 10.5M
+========= ======== ======
+
+Pin Assignment
+--------------
+
+The K210 contains a Fully Programmable I/O Array (FPIOA), which can remap any of
+its 256 input functions to any any of 48 output pins. The following table has
+the default pin assignments for the BitM.
+
+===== ========== =======
+Pin   Function   Comment
+===== ========== =======
+IO_0  JTAG_TCLK
+IO_1  JTAG_TDI
+IO_2  JTAG_TMS
+IO_3  JTAG_TDO
+IO_4  UARTHS_RX
+IO_5  UARTHS_TX
+IO_6  GPIOHS_1
+IO_7  GPIOHS_2
+IO_8  GPIO_0
+IO_9  GPIO_1
+IO_10 GPIO_2
+IO_11 GPIO_3
+IO_12 GPIO_4     Green LED
+IO_13 GPIO_5     Red LED
+IO_14 GPIO_6     Blue LED
+IO_15 GPIO_7
+IO_16 GPIOHS_0   ISP
+IO_17 GPIOHS_3
+IO_18 I2S0_SCLK  MIC CLK
+IO_19 I2S0_WS    MIC WS
+IO_20 I2S0_IN_D0 MIC SD
+IO_21 GPIOHS_4
+IO_22 GPIOHS_5
+IO_23 GPIOHS_6
+IO_24 GPIOHS_7
+IO_25 GPIOHS_8
+IO_26 SPI1_D1    MMC MISO
+IO_27 SPI1_SCLK  MMC CLK
+IO_28 SPI1_D0    MMC MOSI
+IO_29 GPIOHS_31  MMC CS
+IO_30 GPIOHS_9
+IO_31 GPIOHS_10
+IO_32 GPIOHS_11
+IO_33 GPIOHS_12
+IO_34 GPIOHS_13
+IO_35 GPIOHS_14
+IO_36 GPIOHS_28  Panel CS
+IO_37 GPIOHS_29  Panel RST
+IO_38 GPIOHS_30  Panel DC
+IO_39 SPI0_SCK   Panel WR
+IO_40 SCCP_SDA
+IO_41 SCCP_SCLK
+IO_42 DVP_RST
+IO_43 DVP_VSYNC
+IO_44 DVP_PWDN
+IO_45 DVP_HSYNC
+IO_46 DVP_XCLK
+IO_47 DVP_PCLK
+===== ========== =======
+
+Over- and Under-clocking
+------------------------
+
+To change the clock speed of the K210, you will need to enable
+``CONFIG_CLK_K210_SET_RATE`` and edit the board's device tree. To do this, add a
+section to ``arch/riscv/arch/riscv/dts/k210-maix-bit.dts`` like the following:
+
+.. code-block:: dts
+
+    &sysclk {
+ 	assigned-clocks = <&sysclk K210_CLK_PLL0>;
+ 	assigned-clock-rates = <800000000>;
+    };
+
+There are three PLLs on the K210: PLL0 is the parent of most of the components,
+including the CPU and RAM. PLL1 is the parent of the neural network coprocessor.
+PLL2 is the parent of the sound processing devices. Note that child clocks of
+PLL0 and PLL2 run at *half* the speed of the PLLs. For example, if PLL0 is
+running at 800 MHz, then the CPU will run at 400 MHz. This is the example given
+above. The CPU can be overclocked to around 600 MHz, and underclocked to 26 MHz.
+
+It is possible to set PLL2's parent to PLL0. The plls are more accurate when
+converting between similar frequencies. This makes it easier to get an accurate
+frequency for I2S. As an example, consider sampling an I2S device at 44.1 kHz.
+On this device, the I2S serial clock runs at 64 times the sample rate.
+Therefore, we would like to run PLL2 at an even multiple of 2.8224 MHz. If
+PLL2's parent is IN0, we could use a frequency of 390 MHz (the same as the CPU's
+default speed).  Dividing by 138 yields a serial clock of about 2.8261 MHz. This
+results in a sample rate of 44.158 kHz---around 50 Hz or .1% too fast. If,
+instead, we set PLL2's parent to PLL1 running at 390 MHz, and request a rate of
+2.8224 * 136 = 383.8464 MHz, the achieved rate is 383.90625 MHz. Dividing by 136
+yields a serial clock of about 2.8228 MHz. This results in a sample rate of
+44.107 kHz---just 7 Hz or .02% too fast. This configuration is shown in the
+following example:
+
+.. code-block:: dts
+
+    &sysclk {
+ 	assigned-clocks = <&sysclk K210_CLK_PLL1>, <&sysclk K210_CLK_PLL2>;
+	assigned-clock-parents = <0>, <&sysclk K210_CLK_PLL1>;
+ 	assigned-clock-rates = <390000000>, <383846400>;
+    };
+
+There are a couple of quirks to the PLLs. First, there are more frequency ratios
+just above and below 1.0, but there is a small gap around 1.0. To be explicit,
+if the input frequency is 100 MHz, it would be impossible to have an output of
+99 or 101 MHz. In addition, there is a maximum frequency for the internal VCO,
+so higher input/output frequencies will be less accurate than lower ones.
diff --git a/include/configs/sipeed-maix.h b/include/configs/sipeed-maix.h
new file mode 100644
index 0000000000..a46473fc78
--- /dev/null
+++ b/include/configs/sipeed-maix.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/*
+ * Copyright (C) 2019-20 Sean Anderson <seanga2@gmail.com>
+ */
+
+#ifndef CONFIGS_SIPEED_MAIX_H
+#define CONFIGS_SIPEED_MAIX_H
+
+#include <linux/sizes.h>
+
+#define CONFIG_SYS_LOAD_ADDR 0x80000000
+/* Start just below the second bank so we don't clobber it during reloc */
+#define CONFIG_SYS_INIT_SP_ADDR 0x803FFFFF
+#define CONFIG_SYS_MALLOC_LEN SZ_128K
+#define CONFIG_SYS_CACHELINE_SIZE 64
+
+#define CONFIG_SYS_SDRAM_BASE 0x80000000
+/* Don't relocate into AI ram since it isn't set up yet */
+#define CONFIG_SYS_SDRAM_SIZE (SZ_4M + SZ_2M)
+
+/* For early init */
+#define K210_SYSCTL_BASE 0x50440000
+
+#endif /* CONFIGS_SIPEED_MAIX_H */
-- 
2.25.0