[Bug 284042] FTDI: UART breaks JTAG
- In reply to: bugzilla-noreply_a_freebsd.org: "[Bug 284042] FTDI: UART breaks JTAG"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Tue, 14 Jan 2025 22:37:28 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=284042 --- Comment #7 from Tomasz "CeDeROM" CEDRO <tomek@cedro.info> --- Very cool that modern uftdi can disable JATG port not to be grabbed by ucom it was not there back then. From my initial report it seems it was working as ucom then jtag then ucom then jtag.. so probably a bug in uftdi / ucom :-( According to FT2232H datasheet it is possible to configure two channels independently so one works as UART and the other one works as MPSSE in theory they should not disturb each other. https://ftdichip.com/wp-content/uploads/2024/09/DS_FT2232H.pdf Regarding USB-Serial-JTAG the ESP32-C3 and ESP-C6 (RISC-V CPU) indeed has some minimalistic implementation. But if you look at ESP32-S3 TRM Chapter 33 Page 1247: > The CPU JTAG signals can be routed to the USB Serial/JTAG Controller or external GPIO pads using eFuses > and when the user program has started, software control as well. At that time, the JTAG signals from the USB > Serial/JTAG can also be routed to the GPIO matrix. This allows debugging a secondary SoC via JTAG using the > ESP32-S3 USB Serial/JTAG Controller. It would not be RISC-V CPU but the probe can be built that way utilizing ESP's JTAG Command Processor. What is more both channels use CDC-ACM so it should be also possible to expose this simple serial transport over BLE or WIFI. Wish I had time to play around with it. Unfortunately I could not find anyone interested in funding this kind of research. Maybe you can make it happen :-) https://www.espressif.com/sites/default/files/documentation/esp32-s3_technical_reference_manual_en.pdf -- You are receiving this mail because: You are the assignee for the bug.