Angus Lees

> I think I could hack this right now, by having a beefy config file that contained all expected dependencies from the upstream, locking that config, and then passing it...

> Really, the end-goal is to get everyone to run a single, shared ipfs daemon on every machine. To get there, we're going to need to make it possible to...

> Maybe you don't want a toolchain, you want a direct (target configured) dependency on the tools, so that they can be bundled into your rule's output? I agree, but...

> So part of the solution is probably to be more careful of the target/exec distinction, here. I think we're talking past each other. I (believe I am) familiar with...

> This looks very related to https://github.com/bazelbuild/bazel/discussions/21805 Related, but not the same - just to prevent an overzealous close-as-duplicate. I agree `--run_under` (and `run` generally) should be built for host...

Oh wow - I remember considering this, but I think I must have just incorrectly concluded that transitioning `--extra_execution_platforms` wouldn't work, without trying. Fwiw on bazel 6.4.0, I needed to...

> This can be solved by defining two toolchain_type for the toolchain Yes. I tried to respond to this solution in my original issue description: > One onerous current workaround...

> Not very toolchain, only those who want to be used in cfg=target As described [earlier](https://github.com/bazelbuild/bazel/issues/19645#issuecomment-2144311028) in this github comment thread, the toolchain _author_ can't know if their toolchain is...

@yangxuanjia: Question: Was there much analysis behind the decision to use a 1k batch size? In particular, I'm wondering what happens if we stream each result individually (sort of equivalent...