jingham added inline comments.
================
Comment at: lldb/include/lldb/Core/Debugger.h:435-436
+ ReportInterruption(InterruptionReport(cur_func,
+ llvm::formatv(formatv,
+ std::forward<Args>(args)...)));
+ }
----------------
mib wrote:
> If you use `LLVM_PRETTY_FUNCTION` you won't need this.
I'm not sure I understand how you want to use LLVM_PRETTY_FUNCTION. From what
I can tell, LLVM_PRETTY_FUNCTION prints something that looks like the function
invocation as written. But ReportInterruption and the associated macros
provide a formatv API.
================
Comment at: lldb/source/Target/TargetList.cpp:518
"target already exists it the list");
+ UnregisterInProcessTarget(target_sp.get());
m_target_list.push_back(std::move(target_sp));
----------------
bulbazord wrote:
> I'm somewhat puzzled by this. Why are we unregistering the target in
> `AddTargetInternal` when we're registering it in `CreateTargetInternal`?
> Maybe I don't understand the intent behind
> `{Register,Unregister}InProcessTarget`.
The point behind this is someone says "make me a target" and then in the
process of making the target, the Target code does something slow (e.g. look
for external debug files) that we want to interrupt. So we need a way for the
debugger to find the "in the process of being made" Targets.
"AddTargetInternal" is the API where the target gets added to the Debugger's
TargetList, so after that method has run we will find the target in the regular
TargetList. But between CreateTargetInternal and AddTargetInternal, that
target isn't stored anywhere that the Debugger knows how to query. I added
this list to hold the targets that are in the process of getting made, before
they get added to the Debugger's TargetList.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D154542/new/
https://reviews.llvm.org/D154542
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
