Forum Replies Created
-
Posts
-
Hi,
No problem. Based on your description, this is most likely caused by some setting specific to your project files.
First of all, please try the recently released VisualGDB 5.4 Preview1: https://sysprogs.com/files/visualgdb/VisualGDB-5.4-preview1.msi
If the problem still persists, please try either attaching one of your project files (.vcxproj, .filters, .vgdbsettings, .props files should be sufficient, you don’t need to attach the actual source code) or attach another VS instance to the primary one (in the Managed Debug mode) and try reproducing the problem. The outer VS instance should catch the crash and display detailed information about it.
Hi,
Yes, normally just running ESPImageTool.exe <ELF file> [optional FLASH settings] should do the job.
Please feel free to look through this tutorial to get a basic understanding of ESP8266 OTA with VisualGDB: https://visualgdb.com/tutorials/esp8266/ota/
Hi,
This looks like a limitation of newlib-nano. Please follow the link below for known workarounds: https://github.com/RIOT-OS/RIOT/issues/1891
Hi,
OK, we have released v5.4 Preview 1 that supports the new advanced ESP-IDF project subsystem. You can download the build here. You can read the announcement with the main highlights of the new subsystem here: https://sysprogs.com/w/announcing-visualgdb-5-4-preview-1/.
Hi,
First of all, sorry about the broken ‘find all’ function. The internal VS implementation of this function is hardcoded to use files on the Windows machine only, so it indeed doesn’t work with projects accessed fully via SSH yet (we will eventually provide our own equivalent of this function). The “find all references”, code map and other functions should work as expected though.
With eigen, please try locating a specific .cpp file that shows the missing header files, locate it in the IntelliSense Diagnostics Console -> Project View and click on it (the gear icon means that the file has overridden build arguments, the “X” icon means that the file does not participate in build and doesn’t indicate an error). Once you click at that file, VisualGDB will display the CFLAGS used for that file at the bottom of the diagnostics console window.
Please check if the CFLAGS shown there mention the directory where the missing headers are located. If you are not sure, please let us know the details (full directory path and exact CFLAGS).
Hi,
Sorry for the confusion. The Embedded edition indeed doesn’t support the Linux-based targets as shown on the feature matrix on the buy page.
If you have purchased the Embedded edition within the last month, you can upgrade it to Linux edition by paying just the price difference here: http://sysprogs.com/splm/mykey (otherwise maintenance period extension surcharge will apply).
Hi,
No problem, please try this build: http://sysprogs.com/files/tmp/VisualGDB-5.4.1.2095.msi. You will need to change any setting (e.g. port number) on the Settings page in order to save the settings with all variables present (newly created projects won’t need this as long as you are using the new build).
Hi,
The add_subdirectory() command should not normally affect the include directories, it only tells CMake to look for CMakeLists.txt in a specified location.
The behavior you are describing might also be caused by disabled “aggressive completion” mode. If auto-completion works when you press Ctrl-Space, please click at the lightning icon symbol in the completion popup to enable the aggressive mode.
Unfortunately it’s hard to say what is going on with the diagnostics console without a screenshot. If you could attach one, we could help you fix this.
Hi,
Sorry for the confusion. If the -ggdb flag is present, the debug symbols could be accidentally removed from the final executable by using the ‘strip’ command. We have published a detailed tutorial showing how to diagnose symbol-related problems here: https://visualgdb.com/tutorials/linux/symbols/
Please feel free to follow it and let us know if you have any questions. If you are using a cross-compiler, the equivalent of the objdump tool is arm-linux-gnueabihf-objdump.exe (you would need to search its output for “.debug_xxx” sections manually as grep is not available on Windows unless you have Cygwin installed).
Hi,
Please try reloading the remote directories on the “IntelliSense Directories” page (not IntelliSense Settings). If this doesn’t help, please locate a specific header file that is not found and check if its directory is listed under Clang IntelliSense Diagnostics Console -> Project View->Default CFLAGS (the corresponding command-line option should look like -I<directory>).
Hi,
Thanks for checking this. Soldering the wires to the board could be risky and might end up breaking the board (or even your PC), so we cannot advise doing it unless you are OK with the risk.
Another option would be to try using PyOCD instead of OpenOCD. PyOCD is a part of mbed and should be well-integrated with mbed boards and debug interfaces. We have published a detailed tutorial showing how to setup PyOCD with VisualGDB: https://visualgdb.com/tutorials/arm/pyocd/
Hi,
This looks like something specific to the OpenOCD FLASH erasing logic. Although this is normally not covered by our support (please consider using Segger J-Link if you are looking for a debug probe that just works out-of-the-box and comes with its own support), we have published a detailed tutorial showing how to build our OpenOCD fork from scratch and debug it with VisualGDB: https://visualgdb.com/tutorials/arm/openocd/build/
Please ensure you use the following VisualGDB build: http://sysprogs.com/files/tmp/VisualGDB-5.4.1.2094.msi
This should help understand why OpenOCD decides to merge the 2 erase regions and tweak it if necessary.
Hi,
Looks like your project is built without debugging symbols. Please ensure you specify the “-ggdb” flag in the CFLAGS/CXXFLAGS.
Hi,
This looks like some sort of a wiring problem. Please double-check the power and reset connections and try using a different example (e.g. LEDBlink) to see the exact moment where the board starts/stops working.
