3 ms·
Do I understand correctly that the load-time code specialization described in this presentation is not inherent to the Mill hardware, but instead merely a propo
by lealanko 11y ago
Do I understand correctly that the load-time code specialization described in this presentation is not inherent to the Mill hardware, but instead merely a proposed OS-level mechanism for supporting binaries that are portable across the model family? Could a Mill-targeting compiler choose to bypass this mechanism and directly output model-specific concrete code?
- wtallis 11y agoThe specialization isn't necessarily load-time. I think they're envisioning it as most likely being an install-time process, so that distributed code is still portable but they don't have to worry quite so hard about the latency of the specializer. Ahead-of-time specialization will still need to be possible for creating things like bootloaders and OS install media. I think it's safe to say that the future will have a lot of fat binaries.