Re: [RFA] Reverse Debugging, 1/5

[Michael Snyder <msnyder@vmware.com>]

Joel Brobecker wrote:
>>> My own view was a little simpler: Delete the target_set_execdir method,
>> I'm guessing you meant "get" here?
> 
> Ooops! yes, that' right.
> 
>>> and replace all the calls with a reference to the infrun global. I would
>>> keep the target_set_execdir more or less as is; otherwise, you'll need
>>> some kind of observer to notice when the execdir changes. The
>>> "to_can_go_backwards" is an interesting idea, but in my opinion only
>>> makes sense if the target_set_execdir method is removed.  Otherwise,
>>> we can treat target_set_execdir == NULL as cannot-go-backwards.
>> OK, so you're saying that "target_set_execdir" will set the
>> global infrun variable, not a target-defined variable?
> 
> Not quite, actually. Roughly, when the user changes the exec direction,
> we call target_set_execdir and, if the call succeeded, then update the
> direction known in infrun. Or perhaps you or others have better ideas?

Just, well, then target_set_execdir does not actually
set the exec direction -- instead it sort of tells you
if it's *ok* to set the exec direction -- which is what
I would think of as target_can_reverse.

I'm putting this together, and will have a new patch submission shortly.