I was looking into PrintAsync because Pavel brought it up in D121500. The whole point seems to be to let the IOHandler deal with printing the output. In that context it didn't make much sense (to me) that this is using the debugger's output and error stream rather than the one from the IOHandler.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
Comment Actions
In that context it didn't make much sense (to me) that this is using the debugger's output and error stream rather than the one from the IOHandler.
I don't know why this is set up the way it is, but I would be inclined to agree with you.
Comment Actions
(I suspect this has something to do with the ability to override the output of a particular iohandler in the stack. The idea may have been that the iohandler prints whereever it prints, but the asynchronous output always goes to the main output. But I don't know if we make use of this functionality anywhere.)