[Fixed] 2020 Jan. 14 Zephyr > STM32F769I-DISCO > microSD + FatFS > failed in "samples/subsys/fs/fat_fs" > CMD0 and 0x01

Environment

Ubuntu 18.04 LTS
Board: STM32L476 Nucleo_64 (hereafter STM32L476)
Board: STM32F769 Discovery Kit (hereafter STM32F769)
Zephyr 2.1.99
microSD module: CK-40 (SUNHAYATO)
logic monitor: Analog Discovery 2 (hereafter、AD2)

Summary

• Project: samples/subsys/fs/fat_fs
• failed in either of
  • 1. sending CMD0 (0x40 ....) : send 0x20 ... instead
  • 1. receiving 0x01 : receives 0xC0 or 0xE0

How to reproduce

Board

• STM32F769 Discovery Kit
  • Supported Board information at Zephyr
• microSD CK-40
  • Connector conversion board for MicroSD CK-40

Connection between CK-40 (microSD) and STM32F769

• Left: CK-40, Right: STM32F769
  • DO(MISO) --------------- CN9:5 (PB14)
  • VSS -------------------- CN9:7 (GND)
  • CLK -------------------- CN9:6 (PA12)
  • VDD -------------------- CN11:4 (3V3)
  • CMD/DI(MOSI) ------------ CN9:4 (PB15)
  • CS------4.7k(up) ------ CN9:3 (PA11)　<<< Mistake!
  • CS ------ 4.7k(up) ------ CN9:2 (PH6)　<<< Correct (March. 12, 2020)

prj.conf

I used SPI2 for STM32F769.

prj.conf

CONFIG_SPI=y
# CONFIG_SPI_1=y
CONFIG_SPI_2=y
CONFIG_DISK_ACCESS=y
CONFIG_DISK_ACCESS_SDHC=y
CONFIG_DISK_ACCESS_SPI_SDHC=y
CONFIG_LOG=y
CONFIG_FILE_SYSTEM=y
CONFIG_FAT_FILESYSTEM_ELM=y
CONFIG_PRINTK=y

overlay

I prepared the overlay file for STM32F769 as stm32f769i_disco.overlay.

stm32f769i_disco.overlay

/*
2019-12-15
branched from [nrf52840_blip.overlay]
*/

/*
 * Copyright (c) 2019 Tavish Naruka <[e-mail]>
 *
 * SPDX-License-Identifier: Apache-2.0
 */

&spi2 {
        status = "okay";
        cs-gpios = <&gpioh 6 0>;

        sdhc0: sdhc@0 {
                compatible = "zephyr,mmc-spi-slot";
                reg = <0>;
                status = "okay";
                label = "SDHC0";
                spi-max-frequency = <24000000>;
        };
};

build and run

$ west build -p auto -b stm32f769i_disco samples/subsys/fs/fat_fs/

$ west flash
$ west flash

(The second of the west flash is just in case).

$ west debug
[here I set up the logic monitor to start recording]
(gdb) c

Result > Case A (Failue in sending CMD0)

The CMD0 command is garbled as 0x20 0x00 0x00 0x00 0x00 0x4A instead of 0x40 0x00 0x00 0x00 0x00 0x94.

Result > Case B (Success in sending CMD0, but failed in receiving 0x01)

The CMD0 command is sent, but received 0xE0 and 0x7F instead of 0x01.

Result > Case C (Success in sending CMD0, but failed in receiving 0x01)

The CMD0 command is sent, but received 0xC0 instead of 0x01.

Appendix

with STM32L476

With STM32L476 and CK-40, the sample project works fine.

• STM32L476 + SPI1: O.K.
• STM32L476 + SPI2: O.K.
• STM32F769 + SPI2: N.G.

logic with STM32L476 (O.K.)

80 Clocks, CMD0 and receiving 0x01

Related forum post

• Failure of CMD0 sending using STM32F769 + SPI2 for [samples/subsys/fs/fat_fs]

I have posted the question on the forum. Then I made this article because the forum post is not good to read through.

Fixed (2020-03-17)