Mooncake adds MUSA GPU support to unblock end-to-end tests on non-NVIDIA hardware
teortaxesTex · x · 2026-07-25
Mooncake adds MUSA build support to unblock non-NVIDIA testing
A GitHub PR adds MUSA GPU support to the Mooncake PG (ProcessGroup) module. The key reason is practical: without a working PG backend on MUSA, EP tests could only use a fallback path, which prevented meaningful validation on non-NVIDIA hardware.
The post also explains several compiler constraints that forced a refactor:
- mcc cannot parse PyTorch headers.
- It cannot handle c10d enum types like OpType or ReduceOp::RedOpType.
- Host code cannot use <<<>>> kernel launch syntax.
- MUSA lacks CUDA Graph support, so those paths must be #ifdef-ed out.
- mtbfloat16 is missing arithmetic operators, so reductions need a float accumulator.
More from Infra
- Alchemy adds effect/Workflow support across Durable Objects, Workflows and Lambda Durable Functions — samgoodwin89 · 2026-10-27
- Alchemy's Lambda Web Functions run plain Node web servers on Lambda, no OpenNext needed — samgoodwin89 · 2026-10-27
- Foldkit Adds SSR Support With a Few Lines of Config on Cloudflare and AWS — samgoodwin89 · 2026-10-27
- Alchemy Cuts Cloudflare Bindings CPU Time by Up to 166× Via Profiling — samgoodwin89 · 2026-10-27
- Importing all 438 AWS services now takes 13.9 MB and 0.14s, down from 227 MB and 1.61s — samgoodwin89 · 2026-10-27
- Alchemy makes all ~2,850 AWS bindings work from any cloud, not just S3/Lambda/DynamoDB — samgoodwin89 · 2026-10-27