TRUTH SOCIAL:根据@verdacloud最近的GitHub工单,NVLink多播在Blackwell“机密计算”上不被支持,导致SGLang Qwen3.5 397B性能下降61%。NVIDIA的“机密计算”完全是垃圾,此外根据NVIDIA自己的《NVIDIA Secure AI with Blackwell and Hopper GPUs》白皮书,Hopper的机密计算也存在完全未加密的NVLink。
TRUTH SOCIAL: NVLink multicast is not supported on Blackwell "Confidential Computing" leading to 61% performance regression on SGLang Qwen3.5 397B according to @verdacloud 's recent github ticket. NVIDIA's "Confidential Computing" is complete slop as in addition Hopper's confidential computing had fully unencrypted NVLink according to NVIDIA's own "NVIDIA Secure AI with Blackwell and Hopper GPUs" Whitepaper.