Change8

Migrating to llama.cpp b9859

Version b9859 introduces 1 breaking change. This guide details how to update your code.

Released: 7/1/2026

1
Breaking Changes
1
Migration Steps
4
Affected Symbols

⚠️ Check Your Code

If you use any of these symbols, you need to read this guide:

openclggml-backend-dllibdl.hget_adreno_bin_kernel_func_t

Breaking Changes

Issue #1

The OpenCL backend now loads precompiled binary kernels from a library. This required adding `libdl.h` to break a cyclic dependency between `ggml-opencl` and `ggml-backend-dl` (which depends on backend libs). If you rely on specific kernel loading behavior, ensure the new library loading mechanism works correctly in your setup.

Migration Steps

  1. 1
    If you are using the OpenCL backend, be aware that kernel loading behavior has changed to use precompiled binary libraries.

Release Summary

This release introduces the capability to load precompiled binary kernels for the OpenCL backend, resolving a dependency cycle in the process. It also provides numerous precompiled binaries across different operating systems and hardware configurations.

Need More Details?

View the full release notes and all changes for llama.cpp b9859.

View Full Changelog