Conversation
..not that the wrong one exists for a given build, but..
plohrmann
left a comment
There was a problem hiding this comment.
Aside from the one comment, looks good to me. Since the behavior of vogleditor would change slightly (although I doubt anyone will care about the limitation), can you add add a line to the readme to identify the change in behavior?
|
|
||
| QString cmdLine = m_pLaunchTracerDialog->get_command_line(); | ||
| QProcessEnvironment env = m_pLaunchTracerDialog->get_process_environment(); | ||
| QProcessEnvironment env = m_pLaunchTracerDialog->get_process_environment (sizeof (void *) < 8); |
There was a problem hiding this comment.
Not that it's particularly common, but this means a 64-bit vogleditor could not launch and trace a 32-bit application, because you're only going to inject the 64-bit tracer.
There was a problem hiding this comment.
Absolutely. The problem is, leaving this as-is was breaking the tracing in my case.
There was a problem hiding this comment.
The true answer to this question, IMO, is figuring out how to build a separate, 32-bit tracer using Nix.
There was a problem hiding this comment.
@plohrmann, I have addressed your suggestion (I think)
|
@plohrmann, done! |
c617f4c to
7c12134
Compare
| vogleditor_output_message(msg.toStdString().c_str()); | ||
| vogl_printf(msg.toStdString().c_str()); | ||
| vogl_printf("%s", msg.toStdString().c_str()); | ||
| vogl_printf("\n"); |
There was a problem hiding this comment.
can be merged with the previous line "%s\n"
This is an attempt to remedy #217.