6 Mass Storage Device
The Nano Debugger includes a simple Mass Storage Device implementation, which is accessible for read/write operations via the host operating system to which it is connected.
- Read access to basic text and HTML files for detailed kit information and support
- Write access for programming Intel® HEX and UF2 formatted files into the target device’s memory
- Write access for simple text files for utility purposes
6.1 Mass Storage Device Implementation
The Nano Debugger implements a highly optimized variant of the FAT12 file system with several limitations, partly due to the nature of FAT12 itself and optimizations made to fulfill its purpose for its embedded application.
The Curiosity Nano USB device is USB Chapter 9-compliant as a mass storage device but does not, in any way, fulfill the expectations of a general purpose mass storage device. This behavior is intentional.
When using the Windows operating system, the Nano Debugger enumerates as a Curiosity Nano USB Device found in the disk drives section of the device manager. The CURIOSITY drive appears in the file manager and claims the following available drive letter in the system.
The CURIOSITY drive contains approximately 1 MB of free space and does not reflect the target device’s Flash size. When programming an Intel HEX or UF2 file, the binary data are encoded in ASCII with metadata providing significant overhead. Therefore, 1 MB is an arbitrarily chosen value for the disk size.
It is not possible to format the CURIOSITY drive. When programming a file to the target, the filename may appear in the disk directory listing; however, this is merely the operating system’s view of the directory, which in reality has not been updated. It is not possible to read out the file contents. Removing and replugging the board will return the file system to its original state, but the target will still contain the previously programmed application.
Copy a text file starting with “CMD:ERASE” onto the disk to erase the target device.
By default, the CURIOSITY drive contains several read-only files used for generating icons, reporting status, and linking to further information:
AUTORUN.ICO– icon file for the Microchip logoAUTORUN.INF– system file required for Windows Explorer to show the icon fileCLICK-ME.HTM– redirect to a kit-specific web demo applicationKIT-INFO.HTM– redirect to the development board websiteKIT-INFO.TXT– a text file containing details about the board’s debugger firmware version, board name, USB serial number, device, and drag-and-drop supportPUBKEY.TXT– a text file containing the public key for data encryptionSTATUS.TXT– a text file containing the programming status of the board
STATUS.TXT. However, the contents may not reflect the correct status because the operating system may cache the file.6.2 Fuse/Configuration Bytes/Words
Fuse Bytes (AVR® MCU Targets)
When performing drag-and-drop programming, the Nano Debugger masks out any fuse bits that attempt to disable Unified Program and Debug Interface (UPDI), meaning that the UPDI pin cannot be used in its Reset or GPIO modes. Selecting one of the alternative functions of the UPDI pin will render the device inaccessible without using an external debugger capable of high-voltage UPDI activation.
Configuration Bytes/Words (PIC® MCU Targets)
When performing drag-and-drop programming, the debugger masks out any configuration bytes or words that attempt to disable the ICSP interface, or other One-Time Programmable (OTP) configurations.
6.3 Limitations of Drag-and-Drop Programming
ARM Cortex Devices
Drag-and-drop programming is not supported on ARM Cortex devices. Use one of the supported IDEs or command-line utilities for programming.
Lock Bits
Lock bits included in the hex file will be ignored when using drag-and-drop programming. To program lock bits, use one of the supported IDEs.
Enabling CRC Check in Fuses
It is not advisable to enable the CRC check fuses in the target device when using drag-and-drop programming because a subsequent chip erase (which does not affect fuse bits) will cause a CRC mismatch, and the application will fail to boot. A chip erase must be performed using a supported IDE, which automatically clears the CRC fuses after erasing to recover the target from this state.
6.4 Special Commands
Several utility commands are supported by copying text files to the mass storage disk. The filename or extension is irrelevant – the command handler reacts to content only.
| Command Content | Description |
|---|---|
CMD:ERASE | Executes a target chip erase |
CMD:SEND_UART= | Sends a string of characters to the CDC UART. See Advanced Use. |
| Sends a string of characters to the CDC UART at the specified baud rate. Note that only the baud rates explicitly specified here are supported. See Advanced Use (Debugger firmware v1.25 or newer.) |
CMD:RESET | Resets the target device by entering Programming mode and then exiting Programming mode immediately afterwards. The exact timing may vary depending on the programming interface of the target device. (Debugger firmware v1.25 or newer.) |
CMD:POWERTOGGLE | Powers down the target and restores it after a 100 ms delay. If external power is provided, this command has no effect. (Debugger firmware v1.25 or newer.) |
CMD:0V | Powers down the target device by disabling the target supply regulator. If external power is provided, this has no effect. (Debugger firmware v1.25 or newer.) |
CMD:1V8 | Sets the target voltage to 1.8V. If using external power, this has no effect. (Debugger firmware v1.25 or newer.) |
CMD:3V3 | Sets the target voltage to 3.3V. If using external power, this has no effect. (Debugger firmware v1.25 or newer.) |
CMD:5V0 | Sets the target voltage to 5.0V. If using external power, this command has no effect. (Debugger firmware v1.25 or newer.) |
6.5 Drag-and-Drop Programming Using UF2 Format
UF2 Format for Drag-and-Drop
Drag-and-Drop programming provides a simple mechanism for programming the non-volatile memories of the microcontroller on a development kit. This is typically done using the Intel® HEX format, which includes the necessary addresses and segmentation information as part of the format. Intel HEX files contain the memory contents encoded as ASCII characters, which must be parsed in strictly sequential order. This means that the host operating system must send the content in the correct sequence. This is not always the case for all variants of all operating systems. The UF2 format was developed by Microsoft as a means to allow memory to be transferred out of sequence. This is achieved by enforcing a fixed block size throughout the entire data transfer path, thereby preventing partial writes.
More information on the UF2 format is available on GitHub.
Generating a UF2 file
The result of a project compilation procedure is typically an Intel hex file, which can be converted into a UF2 file using a post-build step.
An Intel HEX file can be converted into a UF2 file using the pymcuprog command-line interface:
pymcuprog makeuf2 -f app.hex --uf2file app.uf2pymcuprog makeuf2 -f ${ImagePath} --uf2file ${ImageDir}\${ProjectName}.X.${IMAGE_TYPE}.uf2Each time the application is built, the produced Intel hex file will automatically be converted to a UF2 file with the same filename but with a .uf2 file extension.