My bad, should have been 1.  Will you check in this fix?  I didn’t actually 
have access to one of the “bad” systems that don’t return a stop reason on 
first attach, on our end it was some weird embedded stub.  Glad FreeBSD tests 
this!

Jim

On Jan 17, 2014, at 7:35 AM, Ed Maste <[email protected]> wrote:

> On 14 January 2014 22:32, Jim Ingham <[email protected]> wrote:
>> Author: jingham
>> Date: Tue Jan 14 21:32:42 2014
>> New Revision: 199290
>> 
>> URL: http://llvm.org/viewvc/llvm-project?rev=199290&view=rev
>> Log:
>> Fix a bug where if we stop but nobody says there was a reason for the stop, 
>> we would return
>> control to the user anyway.
> 
> After this change LLDB fails for me on FreeBSD:
> 
> joule% bin/lldb /usr/bin/true
> Current executable set to '/usr/bin/true' (x86_64).
> (lldb) run
> error: initial process state wasn't stopped: exited
> 
> Is the first stop ID not 1 (instead of 0)?  That's what I see on
> FreeBSD and in my Debian VM.  My guess is that we have the issue you
> describe in the commit message and I'll investigate, but changing the
> test to "GetStopID() > 1" restores the existing behaviour for me.


_______________________________________________
lldb-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/lldb-commits

Reply via email to