Re: [RFA] Stop infrun from tracking breakpoint insertion status.

[Vladimir Prus <ghost@cs.msu.su>]

Ulrich Weigand wrote:

> Vladimir Prus wrote:
> 
>> diff --git a/gdb/breakpoint.c b/gdb/breakpoint.c
>> index e81ec20..ef55a3d 100644
>> --- a/gdb/breakpoint.c
>> +++ b/gdb/breakpoint.c
>> @@ -296,6 +296,8 @@ int breakpoint_count;
>>  /* Pointer to current exception event record */
>>  static struct exception_event_record *current_exception_event;
>>  
>> +static int breakpoints_meant_to_be_inserted_p;
>> +
>>  /* This function returns a pointer to the string representation of the
>>     pathname of the dynamically-linked library that has just been
>>     loaded.
>> @@ -1312,6 +1314,12 @@ remove_breakpoints (void)
>>  }
>>  
>>  int
>> +breakpoints_meant_to_be_inserted (void)
>> +{
>> +  return breakpoints_meant_to_be_inserted_p;
>> +}
>> +
>> +int
>>  remove_hw_watchpoints (void)
>>  {
>>    struct bp_location *b;
> 
> 
> Is the breakpoints_meant_to_be_inserted_p variable actually ever
> set to any non-zero value in this patch?

Oops! It's not, which suggest that the three uses of that function are:
(i) not excercised by gdb testsuite and (ii) they are probably less
important than I originally though. 

I'll put the right assignemnts to breakpoints_meant_to_be_inserted_p
and look at the code in question again. 

Sorry,
Volodya