Saving breakpoints after re-sourcing binary - c++

I have a small work environment related doubt. I am analyzing a binary in LLDB and sometimes, I need to make some changes in the code and re-compile it. And then re-source the new binary into LLDB for further analysis.
Currently, I am doing this
Inside LLDB, use shell <shell-command> to compile the code.
Use file <binary> to reload the binary.
But in this way, I am losing the breakpoints. So, is there any way I can save breakpoints?

Couple of things.
First off, if you are recompiling the binary at the same path you used for your current lldb target, then you don't need to make a new target. lldb will notice the file has changed when you do run read in the new binary & debug information, reset the breakpoints, etc.
But if there are other reasons why you need to make a new target, lldb has breakpoint write and breakpoint read commands that allow you to serialize the breakpoints to a file and then read them back into a new target.

Related

How to compile and execute a stand-alone SML-NJ executable

I have seen one other answer link but what I don't understand is what is basis.cm and what's it's use?
You are asking two questions.
What is basis.cm and what's it's use?
This is the Basis library. It allows the use of built-in functions.
How to compile and execute a stand-alone SML-NJ executable
Assuming you followed Jesper Reenberg's tutorial on how to execute a heap image, the next thing you need in order to have SML/NJ produce a stand-alone executable is to convert this heap image. One should hypothetically be able to do this using heap2exec, a tool that takes the heap image, e.g. the .x86-linux file generated on my system, and generates an .asm file that can be assembled and linked.
Unfortunately, this tool is not very well-maintained, so you have to
Go to the smlnj.org page and fix the download-link by removing 'www.' (this page and the SourceForge page don't contain the same explanations or assumptions about argument count, and neither page's download link work).
Download and extract this tool, and fix the 'build' script so it points to your ml-build tool
Fix the tool's argument use by changing [inf, outf] to [_, inf, outf]
Run ./build which generates 'heap2asm.x86-linux' on my system
For example, in order to generate an .asm file for the heap2asm program itself, run
sml #SMLload heap2asm.x86-linux heap2asm.x86-linux heap2asm.s
At this point, I have unfortunately been unable to produce an executable that works. E.g. if you run gcc -c heap2asm.s and ld heap2asm.o, you get a warning of a missing _start label. The resulting executable segfaults even if you rename the existing _sml_heap_image label to _start. That is, it seems that a piece of entry code that the runtime environment normally delivers is missing here.
At this point, discard SML/NJ and use MLton for producing stand-alone binaries.

How to use Instruments Time Profiler with Executable

I'm trying to use Mac Instruments Time Profiler to optimize my code for building a MandelBox. I found how to make my executable my target process, but when the program runs, it gives me an error in the Console window saying it cannot find the .txt file associated with the program.
Do I need to tell the profiler where to look to find the file? The text file is already in the same directory as the executable. Any thoughts? Thanks.
This problem is not unique to Instruments. The same thing would presumably happen if your current working directory was something other than the location of your program. For example, if you were to do cd / ; /path/to/yourprogram.
You either need to make your program find its own location and then find its text file as a sibling in the containing directory or take the path of the text file as an argument. Or, you will always have to set the working directory to your program's location before invoking it.
That last approach is an immediate workaround for the problem with Instruments. On the panel where you choose the target executable, you can also configure various parameters, such as arguments, environment variables, and the working directory. Set the working directory to the directory that contains the text file and it should work.

(VS2010 C++) Execute a command every time the program is run?

The IDE I'm using is VS2010 for writing C++
I want to execute the command cmd C:\utilities\unix\tail.exe -f -n15 $(ProjectDir)Log.txt every time the program I'm coding is run from within the IDE. (This command should open a console to track changes made to the file Log.txt)
There are ways to make a command run every time the program is built, but I can't find a way to make a command run whenever the program itself is run, even if it's already built. Where or how might I be able to set that kind of thing up?
I've tried putting $(TargetPath) & C:\utilities\unix\tail.exe -f -n15 $(ProjectDir)Log.txt into the project's Properties->Debugging->Command (TargetPath is the full name of the debug executable) but VS reads the entire thing as a filename and gets confused.
You can create a file run.cmd for example next to the vcxproj file, which would contain:
%1
C:\utilities\unix\tail.exe -f -n15 %2Log.txt
And then in Properties->Debugging->Command you write:
$(ProjectDir)\run.cmd
and in Command Arguments you write:
"$(TargetPath)" "$(ProjectDir)"
I may have misspelled the macros, but you get the idea: it executes first your program and then whatever you want.
Edit: Unfortunately it works only if you start without debugging (Ctrl+F5), because otherwise the debugger tries to attach to run.cmd and complains that the format is unsupported.

How GDB startup files work?

How do I save my current GDB session? and How do I load it again on GDB startup later. There is a brief discussion on .gdbinit in Art of debugging , Chapter 1. But I really don't get the saving part. Is it an autosave?
The .gdbinit file saves some configurations and user-defined commands. When gdb starts, it will try to search .gdbinit file according to some sequences (For the sequences, you can refer https://sourceware.org/gdb/current/onlinedocs/gdb/Mode-Options.html#Mode-Options). For .gdbinit file, you can refer this as an example:https://github.com/gdbinit/Gdbinit/blob/master/gdbinit.
I think you want to store the whole memory dump into the file, then restart gdb, it will reload it. I have search the invoking gdb manual(https://sourceware.org/gdb/current/onlinedocs/gdb/Invoking-GDB.html#Invoking-GDB), but can't find gdb can do it.
Personally, I think this file you request is very similar to core dump file.

breakpoints in GDB

I think this may have been asked earlier but i can't find one that satisfied my requirements.
I am debugging(infact trying to understand) a large project by trying to analyze the code flow in various testsuites. But when i try to set breakpoints at some files, i get the error "no source file named filename found".
So my question is:
Can gdb only accept breakpoints for the source files where the code flow enters.?
Can I set breakpoints over entire lines of a file with something like b filename:*
Will a breakpoint at header file be accepted as header files are just appended at compile time?
Any insights are more than welcome.
Edit
I checked these issues with some hello world code and found same results as pointed out in one of the answers.but my issue in the actual project still remains on. I still get the same error even when i can see the edited output of the same line which is not accepted as a breakpoint.
Edit 2
I got it working but don't understand how and why it works..??
(gdb) b /home/neeraj/BTP/trunk/include/header.h:872
No source file named /home/neeraj/BTP/trunk/include/header.h:872
Make breakpoint pending on future shared library load? (y or [n]) n
(gdb) b /home/neeraj/BTP/trunk/src/driver.cpp:2
Breakpoint 1 at 0x806c61a: file ../../../trunk/src/driver.cpp, line 2.
(gdb) b /home/neeraj/BTP/trunk/include/header.h:872
Breakpoint 2 at 0x8052fa0: file ../../../trunk/include/header.h:872, line 872.
(gdb)
Any deeper insights..?
No.
No.
Yes.
Make sure you compile with -g (debug) option. Make sure the sourcepaths are set correctly. Use directory, show directories and dir commands to see/set.
The other thing to beware of besides shared libraries is that gdb source file names are relative to the directory where the code was compiled. If you haven't compiled with absolute pathnames, use the dir command to add the compilation directory to the list of places gdb searches for source code.
And a hint: I find I am wildly more productive when I use the Data Display Debugger (DDD) graphical front end to gdb.