25-03-2024, 10:48 AM
Oh, I didn't notice you also modified the GDB client. I thought you used it like that: EBlink-standard GNU GDB client-EmBitz.
I am trying to integrate it into my debug workflow. In the final form, I want to have live updates for a few variables/watches while debugging with my chain with EBlink-standard GDB-Debugger.
So my thought was if I use EBlink as the GDB Server, I could use your optimizations to also access the live variables. From the readme.md on Github, it seemed to me as if EBlink would not be only exclusive for EmBitz. Its great that EmBitz is getting all these fancy features. But I don't really need them without an os running on the target and have a few other reasons to go with another IDE for this project.
But as far as I understand, I think your last idea would be a solution. Having working live watches on address level would be the first stage. If this would work, I am confident I can develop my own solution to solve the variable to address resolution in the debugger.
Do you think it is possible to give me a bit of advice on getting the first stage running? I dont want to cause a lot of work for you describing it for me.
I am unsure how you achieved the live variables? Are you working in asynchronous gdb mode, do you set up a second connection to gdb to run monitor commands while the program is running or how do you read memory while running?
I am trying to integrate it into my debug workflow. In the final form, I want to have live updates for a few variables/watches while debugging with my chain with EBlink-standard GDB-Debugger.
So my thought was if I use EBlink as the GDB Server, I could use your optimizations to also access the live variables. From the readme.md on Github, it seemed to me as if EBlink would not be only exclusive for EmBitz. Its great that EmBitz is getting all these fancy features. But I don't really need them without an os running on the target and have a few other reasons to go with another IDE for this project.
But as far as I understand, I think your last idea would be a solution. Having working live watches on address level would be the first stage. If this would work, I am confident I can develop my own solution to solve the variable to address resolution in the debugger.
Do you think it is possible to give me a bit of advice on getting the first stage running? I dont want to cause a lot of work for you describing it for me.
I am unsure how you achieved the live variables? Are you working in asynchronous gdb mode, do you set up a second connection to gdb to run monitor commands while the program is running or how do you read memory while running?