Re: [PATCH v2] Prevent downgrading of hardware watchpoints if possible

 Am Freitag, 21. August 2026 um 19:56:00 MESZ hat Tom Tromey <[email protected]> Folgendes geschrieben:

> >>>>> "Hannes" == Hannes Domani <[email protected]> writes:
> 
> Hannes> The lazy flag of a value can tell us if the value contents are available.
> Hannes> But this could be either because it was simply never actually accessed,
> Hannes> or it tried to be accessed, and failed.
> Hannes> The latter are interesting for watchpoints, the former are not.
> 
> Hannes> Currently it only uses lazy values if they are at the head of the value
> Hannes> chain, but this is fragile logic, and degrades some watchpoints to
> Hannes> software watchpoints where it's actually not necessary.
> 
> Hannes> So this adds a new value flag 'm_fetch_lazy_failed' which tells if the
> Hannes> watchpoint expression actually tried to access the value and failed.
> Hannes> And this is used instead of the value chain location to tell if a lazy
> Hannes> value should be used as a watchpoint location.
> 
> Thanks for doing this.  I think this is ok.
> Approved-By: Tom Tromey <[email protected]>

Pushed, thanks.


Hannes