How to use SDcard adapters via SPI on a STM32?

[rt-2pm2]

I am struggling to have a STM32F4 based board use a micro sdcard adapter over SPI.
I checked that the adapter and the sdcard are compatible and functional (with an arduino).

I started looking at the Disk Access documentation where there is an example of overlay to describe my setup. I took inspiration also looking at the fs_sample example.

My Setup

In my case, the relevant overlay is:

&spi1 {
	status = "okay";
	cs-gpios = <&gpioa 4 GPIO_ACTIVE_LOW>;
	pinctrl-0 = <&spi1_clk_a5 &spi1_miso_a6 &spi1_mosi_a7>;
	pinctrl-names = "default";

	sdhc0: sdhc@0 {
		compatible = "zephyr,sdhc-spi-slot";
		reg = <0>;
		status = "okay";
		mmc {
			compatible = "zephyr,sdmmc-disk";
			disk-name = "SD";
			status = "okay";
		};
		spi-max-frequency = <400000>;
	};
};

&pinctrl {
	/omit-if-no-ref/ spi1_clk_a5: spi1_clk_a5 {
		pinmux = <STM32_PINMUX('A', 5, AF5)>;
	};

	/omit-if-no-ref/ spi1_miso_a6: spi1_miso_a6 {
		pinmux = <STM32_PINMUX('A', 6, AF5)>;
		bias-pull-up;
	};

	/omit-if-no-ref/ spi1_mosi_a7: spi1_mosi_a7 {
		pinmux = <STM32_PINMUX('A', 7, AF5)>;
	};
};

Basically, I set up the pins for the SPI, then I set up the spi interface following the stm32 bindings.

On the KConfig side, the relevant part is as follows:

# ----------------------------------
# SPI / SDCard
# ----------------------------------
CONFIG_SPI=y
CONFIG_SDHC=y
CONFIG_DISK_ACCESS=y
CONFIG_DISK_DRIVERS=y

CONFIG_SPI_SDHC=y

The part of the code where I am trying to initialize the micro sdcard by calling disk_access_ioctl is stripped down to:

int init_sdcard() {
	if (!initialized_) {
		static const char *disk_pdrv = "SD";
		int ret = disk_access_ioctl(disk_pdrv, DISK_IOCTL_CTRL_INIT,  NULL);
		...

Issue

My application does not pass that call and returns the -134 error. Enabling a more detailed debug I spotted:
<dbg> sd: sdmmc_read_scr: ACMD51 failed: -5
I attached a Logic Analyzer to the pins and I spotted a pattern that confirm that error.

• Step 1:
  The Master pulls down the CS and sends a 0xFF 0x40 0x00 0x00 0x00 0x00 0x95
  The Slave replies with a 0x01
• Step 2:
  The Master pulls up and down the CS and sends a 0xFF 0x48 0x00 0x00 0x01 0xAA 0x87
  The Slave replies with a 0x01 0x00 0x00 0x01 0xAA

These steps continue to repeat until the error about 20 times. Reading a bit about the SDCards, those are the CMD0 and CMD8. Then, the MOSI continues to carry other commands until it reaches a 0x4A command, that should CMD10. The MISO replies with a 0x01 and then it gets stuck: The CS remains down, the SCK continue to switch, but both MOSI and MISO remain HIGH, until the error.

I don't understand why the process gets stuck there. Is there somewhere a full working example I can look at?
Thanks in advance for your time ;-).

(edit: typos and update the complete sequence of commands before the freeze)

[rt-2pm2]

Actually, I made a longer recording with the logic analyzer. After about a minute something moves on the line.
There is some movement on the MOSI/MISO which the logic analyzer does not "parse". After that, the CS toggles HIGH, then LOW again, followed by a CMD 0x49 on the MOSI and the 0x01 reply on the MISO.
Then, there is another long pause (60 seconds) with the CS down, which terminates with a CMD 0x4C followed by a 0x05 reply.

Immediately after that I see a 0xC0 on the MOSI. However, this last MOSI CMD is different. The SCK line does not look right (the clock is not regular).

I noticed also another strange fact: the final error is different from the previous one. This time is:

<dbg> sd: sdmmc_read_scr: SD app command failed for SD SCR

I could think it is a cabling/disturbance problem...but then I don't understand why on the Arduino it works flawlessly