vsk added inline comments.
================
Comment at: lldb/source/Target/TargetList.cpp:293
// FIXME: Maybe the dummy target should be per-Debugger
- if (!m_dummy_target_sp || !m_dummy_target_sp->IsValid()) {
+ if (!m_dummy_target_sp) {
ArchSpec arch(Target::GetDefaultArchitecture());
----------------
vsk wrote:
> jingham wrote:
> > JDevlieghere wrote:
> > > I wonder if we gain anything by making this lazy. Maybe it's time to
> > > address that FIXME and always have a single DummyTarget held onto by the
> > > Debugger?
> > I see no reason not to just make a dummy target when you make a debugger.
> > That should be a very cheap operation, and the object itself should not be
> > large.
> >
> > I think for regularities sake it's better to have the TargetList have all
> > the targets, including the Dummy target, rather than have the DummyTarget
> > held specially by the debugger. But it would be easy for the Debugger to
> > make a DummyTarget and inject it into the TargetList as part of its
> > startup. We could even make the straight constructor of TargetList do this
> > job. We copy the debugger TargetList to iterate over it and for similar
> > purposes in a bunch of places, but that goes through the copy constructor
> > so it wouldn't interfere with this.
> You're right, that's even better.
So far, the policy has been to not include the dummy target in the list of
targets (i.e, you can't find it via TargetList::GetTargetAtIndex or
TargetList::GetIndexOfTarget). If, down the line, we need to support that we
can add the functionality though.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D90872/new/
https://reviews.llvm.org/D90872
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
