Problem
The firmware load precedes normally via usb-storage after which bookworm fails to create the serial port /dev/ttyACM0
Equipment
Raspberry Pi 4 running bookworm 6.12.34+rpt-rpi-v8, RP2040 Wireless, PlatformIO remote agent
Details
+ The same RP2040, cable, firmware, and PlatformIO remote setup works as expected on my Raspberry Pi 5 running bookworm
+ The RPi4 will create /dev/ttyACM0 if I first load the firmware using the RPi5 and then move the USB cable to the RPi4
+ RPi4 dmesg -w fails to report the creation of /dev/ttyACM0 unless I preprogram the RP2040 using the RPi5 and move the cable to the RPi4
+ On both the RPi4 and RPi5, dmesg reports the RP2040 serial device is recognized as an Adafruit Feather RP2040 RFM with idVendor=239a and idProduct=812d (My device is a real Pico W although I have used feathers in the past)
Thoughts
+ The problem apparently has something to do with RPi4 bookworm recognizing the RP2040 switching from usb-storage to usb-serial as, if I use the RPi5 to place the RP2040's USB port into serial mode, it works fine on the RPi4. The RPi4 only fails if the RP2040 is connected during the switch.
+ Having a desktop session active, as mentioned in some forums, seems to have no effect. My RPi4 misses the switch from storage to serial.
+ It's hard to imagine no one else has seen this (extensive google and Copilot searches haven't found much) so I'm inclined to believe I have something (bookworm? RP2040? PlatformIO?) fouled-up.
The firmware load precedes normally via usb-storage after which bookworm fails to create the serial port /dev/ttyACM0
Equipment
Raspberry Pi 4 running bookworm 6.12.34+rpt-rpi-v8, RP2040 Wireless, PlatformIO remote agent
Details
+ The same RP2040, cable, firmware, and PlatformIO remote setup works as expected on my Raspberry Pi 5 running bookworm
+ The RPi4 will create /dev/ttyACM0 if I first load the firmware using the RPi5 and then move the USB cable to the RPi4
+ RPi4 dmesg -w fails to report the creation of /dev/ttyACM0 unless I preprogram the RP2040 using the RPi5 and move the cable to the RPi4
+ On both the RPi4 and RPi5, dmesg reports the RP2040 serial device is recognized as an Adafruit Feather RP2040 RFM with idVendor=239a and idProduct=812d (My device is a real Pico W although I have used feathers in the past)
Thoughts
+ The problem apparently has something to do with RPi4 bookworm recognizing the RP2040 switching from usb-storage to usb-serial as, if I use the RPi5 to place the RP2040's USB port into serial mode, it works fine on the RPi4. The RPi4 only fails if the RP2040 is connected during the switch.
+ Having a desktop session active, as mentioned in some forums, seems to have no effect. My RPi4 misses the switch from storage to serial.
+ It's hard to imagine no one else has seen this (extensive google and Copilot searches haven't found much) so I'm inclined to believe I have something (bookworm? RP2040? PlatformIO?) fouled-up.
Statistics: Posted by conr2286 — Sat Aug 23, 2025 11:50 pm