Re: [PATCH 3/3] PR remote/19496, timeout in forking-threads-plus-bkpt

[Don Breazeal <donb@codesourcery.com>]

On 2/1/2016 4:05 AM, Pedro Alves wrote:
> On 01/28/2016 12:48 AM, Don Breazeal wrote:
> 
>>
>> ---
>>  gdb/infrun.c | 8 +++++++-
>>  1 file changed, 7 insertions(+), 1 deletion(-)
>>
>> diff --git a/gdb/infrun.c b/gdb/infrun.c
>> index 15210c9..e324864 100644
>> --- a/gdb/infrun.c
>> +++ b/gdb/infrun.c
>> @@ -5232,7 +5232,13 @@ Cannot fill $_exitsignal with the correct signal number.\n"));
>>  	  ecs->ptid = inferior_ptid;
>>  
>>  	  if (should_resume)
>> -	    keep_going (ecs);
>> +	    {
>> +	      /* Make sure the thread is marked as running.  If the event
>> +		 occurred in the thread before we added it, it may have
>> +		 never been marked running.  */
>> +	      set_running (inferior_ptid, 1);
>> +	      keep_going (ecs);
>> +	    }
>>  	  else
>>  	    stop_waiting (ecs);
>>  	  return;
>>
> 
> Can you check whether this is still needed in current master?
> 
> Specifically, whether a2077e254098 ([PATCH] Fix PR 19461: strange "info thread"
> behavior in non-stop) already fixed this?
> 
> See:
>  https://sourceware.org/ml/gdb-patches/2016-01/msg00446.html
>  https://sourceware.org/bugzilla/show_bug.cgi?id=19461
> 
> Thanks,
> Pedro Alves
> 
Hi Pedro,
It looks like my patch is still needed.  FYI, the values of the
variables used to determine whether to call set_running in your patch
are this, in my failing test case:

follow_child 0
detach_fork 1
non_stop 1
sched_multi 0
target_is_non_stop_p () 1

Thanks,
--Don