[lldb-dev] How to detach from stop mode without causing signal in inferior on Linux?

[Jeffrey Tan via lldb-dev]

Thanks Pavel. I think I know what's going on. We do call SBProcess.Detach()
during exiting python process normally. However, our stop debugging code
seems to only kill python process directly without letting it exit
normally.
Killing python process seems to detach from inferior on Mac but kills
inferior on Linux. But I guess that's because of the implementation
difference, for example, Linux might left an embeded breakpoint in the
inferior process without clean-up which caused the inferior to signal
with "*Trace/breakpoint
trap*".
I will try to send a detach command to python before killing it.

Jeffrey

On Tue, Mar 29, 2016 at 2:08 AM, Pavel Labath <labath at google.com> wrote:

> Hi,
>
> are you doing a SBProcess.Detach() before you kill the debugger? If
> that doesn't help could you send me the gdb-remote packet form
> (execute command: log enable -f /some/file.log gdb-remote packets), so
> I can get a better idea of what is going on.
>
> cheers,
> pl
>
>
> On 29 March 2016 at 00:21, Jeffrey Tan via lldb-dev
> <lldb-dev at lists.llvm.org> wrote:
> > Hi,
> >
> > When user tries to stop debugging, we call SBDebugger.Destroy(), then
> > SBDebugger.Terminate() and exit the python process. This works well on
> mac,
> > but on linux in running mode. However, I found if the debugger hits
> > breakpoint first, then user stops debugging, the inferior is killed with
> > siginal:
> > "Trace/breakpoint trap"
> >
> > I tried to call SBTarget.DeleteAllBreakpoints() first before killing
> > debugger. Still the same result.
> >
> > Jeffrey
> >
> > _______________________________________________
> > lldb-dev mailing list
> > lldb-dev at lists.llvm.org
> > http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-dev/attachments/20160329/db92b118/attachment.html>