strace shows that minicom does in fact print the "Sending BREAK" pop-up, then it calls ioctl TCSBRK, which returns 0 (success) immediately, and minicom tears down the pop-up before it makes it out to HDMI.
Normally the ioctl would take 0.25s. It may be possible for it return immediately and also work, but in this case it is probably not working.
I do get the delay on bookworm x86-64, so this is a difference between Debian 6.1 and RP 6.6 kernels.
I suspect it is not a bug. In lsusb -vd 2e8a:000a we see:The Pico is not setting 0x04 "device supports Send_Break".
I spent way too long trying to work out where in the SDK/TinyUSB this bit of the interface descriptor gets populated, but I cannot find it.
edit: pico-sdk/lib/tinyusb/src/device/usbd.h around line 170:Change 2 to 6, mention break in the comment. lsusb:
Normally the ioctl would take 0.25s. It may be possible for it return immediately and also work, but in this case it is probably not working.
I do get the delay on bookworm x86-64, so this is a difference between Debian 6.1 and RP 6.6 kernels.
I suspect it is not a bug. In lsusb -vd 2e8a:000a we see:
Code:
CDC ACM: bmCapabilities 0x02 line coding and serial stateI spent way too long trying to work out where in the SDK/TinyUSB this bit of the interface descriptor gets populated
edit: pico-sdk/lib/tinyusb/src/device/usbd.h around line 170:
Code:
/* CDC ACM: support line request */\ 4, TUSB_DESC_CS_INTERFACE, CDC_FUNC_DESC_ABSTRACT_CONTROL_MANAGEMENT, 2,\Code:
CDC ACM: bmCapabilities 0x06 sends break line coding and serial stateStatistics: Posted by jojopi — Tue Apr 30, 2024 3:47 am