example stm32f1 usb_serial: unable to enumerate USB device on CH32F103C8T6

[nim65s]

Hi,

On the current main branch, when I run cargo run --bin usb_serial --release in the embassy/examples/stm32f1 folder, it correctly build and flash my bluepill (stm32f103c8) through a stlink v2:

$ cargo run --bin usb_serial --release
    Finished release [optimized + debuginfo] target(s) in 0.25s
     Running `probe-rs run --chip STM32F103C8 target/thumbv7m-none-eabi/release/usb_serial`
     Erasing sectors ✔ [00:00:02] [########################################################] 37.00 KiB/37.00 KiB @ 13.73 KiB/s (eta 0s )
 Programming pages   ✔ [00:00:55] [#########################################################] 37.00 KiB/37.00 KiB @    680 B/s (eta 0s )    Finished in 58.348s
0.000000 TRACE BDCR configured: 00000200
└─ embassy_stm32::rcc::bd::{impl#2}::init @ …/embassy/embassy-stm32/src/fmt.rs:117 
0.000000 DEBUG rcc: Clocks { sys: Hertz(48000000), pclk1: Hertz(24000000), pclk1_tim: Hertz(48000000), pclk2: Hertz(48000000), pclk2_tim: Hertz(48000000), hclk1: Hertz(48000000), adc: Some(Hertz(6000000)), rtc: Some(Hertz(40000)) }
└─ embassy_stm32::rcc::set_freqs @ …/embassy/embassy-stm32/src/fmt.rs:130 
0.000000 INFO  Hello World!
└─ usb_serial::____embassy_main_task::{async_fn#0} @ src/bin/usb_serial.rs:30  

But the device is not recognized on linux. dmesg show the following:

[82148.285782] usb 1-6.4.4.2: new full-speed USB device number 43 using xhci_hcd
[82148.382465] usb 1-6.4.4.2: device descriptor read/64, error -32
[82148.585779] usb 1-6.4.4.2: device descriptor read/64, error -32
[82148.789064] usb 1-6.4.4.2: new full-speed USB device number 44 using xhci_hcd
[82148.885799] usb 1-6.4.4.2: device descriptor read/64, error -32
[82149.089093] usb 1-6.4.4.2: device descriptor read/64, error -32
[82149.196025] usb 1-6.4.4-port2: attempt power cycle
[82149.835745] usb 1-6.4.4.2: new full-speed USB device number 45 using xhci_hcd
[82149.836137] usb 1-6.4.4.2: Device not responding to setup address.
[82150.042773] usb 1-6.4.4.2: Device not responding to setup address.
[82150.249053] usb 1-6.4.4.2: device not accepting address 45, error -71
[82150.345714] usb 1-6.4.4.2: new full-speed USB device number 46 using xhci_hcd
[82150.346162] usb 1-6.4.4.2: Device not responding to setup address.
[82150.552830] usb 1-6.4.4.2: Device not responding to setup address.
[82150.759129] usb 1-6.4.4.2: device not accepting address 46, error -71
[82150.759415] usb 1-6.4.4-port2: unable to enumerate USB device

Am I missing something, or is there a bug here ?

NB: On the same setup, with https://github.com/stm32-rs/stm32f1xx-hal and cargo embed --features stm32f103 --example usb_serial --release --chip STM32F103C8, I correctly get a serial tty:

[83377.084346] usb 1-6.4.4.2: new full-speed USB device number 51 using xhci_hcd
[83377.200551] usb 1-6.4.4.2: New USB device found, idVendor=16c0, idProduct=27dd, bcdDevice= 0.10
[83377.200568] usb 1-6.4.4.2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[83377.200575] usb 1-6.4.4.2: Product: Serial port
[83377.200581] usb 1-6.4.4.2: Manufacturer: Fake company
[83377.200586] usb 1-6.4.4.2: SerialNumber: TEST
[83377.209085] cdc_acm 1-6.4.4.2:1.0: ttyACM0: USB ACM device

[Dirbaio]

can you post full trace-level logs? DEFMT_LOG=trace cargo run --bin usb_serial --release

[nim65s]

Yes:

$ DEFMT_LOG=trace cargo run --bin usb_serial --release
    Finished release [optimized + debuginfo] target(s) in 0.13s
     Running `probe-rs run --chip STM32F103C8 target/thumbv7m-none-eabi/release/usb_serial`
     Erasing sectors ✔ [00:00:02] [########################################################] 37.00 KiB/37.00 KiB @ 14.96 KiB/s (eta 0s )
 Programming pages   ✔ [00:00:55] [##################################################################################################################################################################################################] 37.00 KiB/37.00 KiB @    681 B/s (eta 0s )    Finished in 58.092s
0.000000 TRACE BDCR configured: 00000200
└─ embassy_stm32::rcc::bd::{impl#2}::init @ /home/nim/local/tmp/embassy/embassy-stm32/src/fmt.rs:117 
0.000000 DEBUG rcc: Clocks { sys: Hertz(48000000), pclk1: Hertz(24000000), pclk1_tim: Hertz(48000000), pclk2: Hertz(48000000), pclk2_tim: Hertz(48000000), hclk1: Hertz(48000000), adc: Some(Hertz(6000000)), rtc: Some(Hertz(40000)) }
└─ embassy_stm32::rcc::set_freqs @ /home/nim/local/tmp/embassy/embassy-stm32/src/fmt.rs:130 
0.000000 INFO  Hello World!
└─ usb_serial::____embassy_main_task::{async_fn#0} @ src/bin/usb_serial.rs:30  

[nim65s]

Adding a bit more info!(), I can see it is blocking on the let driver = Driver::new(p.USB, Irqs, p.PA12, p.PA11); line

[nim65s]

And with more info!() in embassy-stm32/src/usb/usb.rs, it is blocking on cortex_m::asm::delay(unsafe { crate::rcc::get_freqs() }.sys.0 / 10);

[Dirbaio]

I just tried the usb_serial example on my bluepill again just in case it had bitrotten, and it works...

Could it be a hardware issue? How are you powering the bluepill, do you have anything else connected?

The setup that works for me is:

• Bluepill USB-C connected to a USB port in the PC
• Bluepill SWDIO, SWCLK, GND to STLINKv2 probe connected to a USB port in the PC.
  • NOT 3v3, NOT 5v. these'll "fight" with the power the bluepill is getting from the usb-c connector.

Could you try that?

[Dirbaio]

I have this bluepill btw. Maybe you have the other version? Both should work though...

[nim65s]

Thanks for trying !

My bluepill has micro-usb, and if I disconnect 3.3v from the stlink "VAPP", I get:

Error: Connecting to the chip was unsuccessful.

Caused by:
    0: An error with the usage of the probe occurred
    1: An error specific to a probe type occurred
    2: Command failed with status JtagGetIdcodeError

About this VAPP connection, datasheet says:

The power supply from the application board is connected to the ST-LINK/V2 debugging and programming board to ensure signal compatibility between the boards.

If I disconnect micro-usb, and only power the bluepill from stlink (with a 3-way cable between bluepill 3v3 and stlink VAPP and VDD), I have the same result (blocking on line cortex_m::asm::delay(unsafe { crate::rcc::get_freqs() }.sys.0 / 10); and 1-6.4.4-port2: unable to enumerate USB device on linux when I plug the micro-usb)

This might be my board. I have that one. They say it is based on STM32F103C8T6, but I can read on the chip that it is CH32F103C8T6.

[Dirbaio]

CH32F103C8T6

the CH32F103 is a chip made by a different company aiming to be register-compatible with the stm32f103. It's known to have slight differences though, which might explain why it doesn't work for you.

Unfortunately Embassy only supports the STM32 chips, we don't actively test others. It's still nice to support other chips, if you figure out what the issue is and send a PR I'll merge it.

[nim65s]

Ok, I understand. I did try a few workarounds on this cortex_m::asm::delay(unsafe { crate::rcc::get_freqs() }.sys.0 / 10); line, but I don't get anything working. I guess I'll close this, and try again on a real STM32 :)

Thanks for your help !