This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: [RFA] Clarify infrun variable naming.


Pierre Muller wrote:

> 
> 
>> +   If we hit a breakpoint or watchpoint, and then continue,
>> +   we need to single step the current thread with breakpoints
>> +   disabled, so that to avoid hitting the same breakpoint or
>> +   watchpoint again.  And we should step just a single
>> +   thread and keep other threads stopped, so that
>> +   other threads don't miss breakpoints while they are removed.
>> +
>> +   So, this variable simultaneously means that we need to single
>> +   step current thread, keep other threads stopped, and that
>> +   breakpoints should be removed while we step.
>   But this is the reason of the failure to catch watchpoints
> that happen at the point where we are just stepping over a breakpoint,
> because we step with the watchpoints disabled.
>   Why don't we enable all break- and watchpoints but the
> ones that do have the same PC we are currently?

Because that's extra work, and I haven't got around to that yet ;-)
In case of watchpoints, you probably meant enabling all watchpoint
at different data address, not PC?
 
>   Enabling at least all watchpoints would fix gdb/38 failure as
> seen in gdb.base/watchpoint.exp where it is noted as a KFAIL.
> 
>   I tried to check this by adding a insert_watchpoint function
> a few days ago, but if you are working on it anyhow,
> and probably master this much better than I do, it would be
> great to solve that issue at the same time.

I plan to address this soon (but as a separate patch).

- Volodya



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]