[Lldb-commits] [PATCH] D148395: [lldb] Unify default/hijack listener between Process{Attach, Launch}Info (NFC)

[Alex Langford via Phabricator via lldb-commits]

bulbazord added a comment.

In D148395#4285017 <https://reviews.llvm.org/D148395#4285017>, @mib wrote:

> In D148395#4270508 <https://reviews.llvm.org/D148395#4270508>, @bulbazord wrote:
>
>> Creating a ProcessAttachInfo from a ProcessLaunchInfo with this change means they'll have different listeners. Is that ProcessLaunchInfo kept around for any other reason? Or is it just made to create the ProcessAttachInfo? This seems like a reasonable move to me, but I'm not sure how LaunchInfo and AttachInfo might interact.
>
> @bulbazord Not sure what you mean ... We need to convert to the `ProcessLaunchInfo` into a `ProcessAttachInfo` when the user ask use to launch a process but we end up asking the platform to do the launch after which we attach to the process. In both cases, we use the default listener (the debugger listener if the user didn't provide a custom listener).

What I'm not sure about is that the `ProcessAttachInfo` constructor we're using takes a `ProcessLaunchInfo` as an argument and fills in its fields with it. It used to be that `ProcessAttachInfo` would get its Listener and HijackListener from the `ProcessLaunchInfo` passed in. Now, they could have different Listeners and HijackListeners. When we create a `ProcessAttachInfo` from a `ProcessLaunchInfo`, is it expected that these 2 things will diverge in that way? Do we use the `ProcessLaunchInfo` that we build the `ProcessAttachInfo` with in any other way?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D148395/new/

https://reviews.llvm.org/D148395