Search the Community
Showing results for tags 'cdc-acm'.
Found 1 result
It seems this is the problem that will never go away. I am asking here because anything on Google appears to be many years old, and I don't know if patches were applied to the Linux code base in that time or not. My Linux version (from /proc/version): "Linux version 4.9.56 (gleepy@clyde) (gcc version 6.3.0 (GCC) ) #1 SMP Sun Oct 15 14:54:08 EDT 2017" as an x86_64 system with no 32 bit libraries anywhere, I am causing the cdc-acm module to be loaded when either Launchpad is attached: two /dev/ttyACMx ports with the MSP-EXP430FR5994, as is expected; one with the '430G2. The '430G2 has its TXD/RXD jumpers oriented horizontally for hardware serial communication. I have no issues with using mspdebug rf2500 to flash and verify any code I compile with MSP430 GCC. When flashing noForth onto the board, there was absolutely no communications with the computer when running minicom 2.7.1 at the recommended settings of 9600 baud, 8 bits, no parity, 1 stop, and no flow control. None. Even when I followed some old advice of using the stty command to set the speed and flush the port immediately after plugging in the Launchpad. Even the "out of the box" firmware on the '430FR5994 will not communicate with the GUI program which is run. No indication of any extant serial ports show up on the selector in the GUI program. This does not appear to be a privilege or permissions issue as even as root, the communications problems exist. Since people seem to be able to use the Launchpad with Linux to invoke serial communications, as it is implied in the TI documentation and what the noForth people say. Therefore, there exists a set of users which do not have this communications problem. Could it be a lost cdc-acm patch that was supposed to work its way to the Linux maintainers and had gotten lost in code checkins?