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: [PATCH] Only use 32-bit thumb-2 breakpoint instruction if necessary


Yao Qi writes:

> Antoine Tremblay <antoine.tremblay@ericsson.com> writes:
>
>> Yes at first I though what's the harm in allowing different breakpoint
>> kinds on an address ?
>
> There is nothing harmful, IMO.  It is a matter of complexity.
>
>>
>> But looking at the code that introduced the check there seems to be complex
>> reasons for that.
>>
>>>I would like to change GDBserver code, and fix the regressions caused
>>> by the change.
>>
>> I'm not sure what you mean ? You want to take over this issue ? (not a
>> problem with me)
>
> No, I have no plan to take this issue over.  I mean, if I were you, I'd
> start from change GDBserver code, fix the regressions, etc.
>
>>
>> I think we may forget this change for 8.0 however... ?
>
> I feel it is difficult to catch 8.0 release, as there are still some open
> problems that we don't have ideas yet.

Since this won't be fixed for 8.0 and that I don't know when I will
rework on this, would it make sense to revert the software single step in
GDBServer ?

Basically as it is now software single step has problems in GDB and
GDBServer where as before 7.11 iirc at least GDBServer was working correctly.

Regards,
Antoine


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