ggml/llama.cpp releases: b10172

TL;DR

ggml-webgpu updates address binding alias issues and overlapping range problems across different architectures, ensuring compatibility and stability. These fixes are crucial for maintaining the project's reliability across various computing environments.

Detailed Summary

The ggml-webgpu project released an update addressing binding alias issues and recurrent-state-rollback tests to support multiple architectures. The broader impact includes improved compatibility across different hardware platforms, enhancing the functionality of the ggml/llama.cpp library for various computing environments.

Key Points

• • Fix some binding alias issues to support all archs
• • Add overlap glu variant to support all archs, fix recurrent-state-rollback test
• • Diagnose bus error on Apple CI