llama_cpp_for_radxa_dragon_.../docs
Chenguang Li 67e3f6f601
CANN: add operator fusion support for ADD + RMS_NORM (#17512)
This commit implements operator fusion for ADD + RMS_NORM operations
in the CANN backend to reduce memory access overhead and improve
performance. The fusion is controlled by the GGML_CANN_OPERATOR_FUSION
environment variable (default: false).

Changes:
- Implement ggml_cann_op_add_rms_norm_fused() using ACLNN AddRmsNorm
- Add ggml_cann_can_fuse() to check fusion eligibility
- Integrate fusion logic into computation graph evaluation
- Add test cases for ADD + RMS_NORM fusion
- Update documentation with new environment variable

The fusion combines ADD and RMS_NORM into a single kernel call,
which is more efficient than executing them separately.
2026-01-05 15:38:18 +08:00
..
android
backend CANN: add operator fusion support for ADD + RMS_NORM (#17512) 2026-01-05 15:38:18 +08:00
development
multimodal
ops metal : add count_equal op (#18314) 2025-12-31 10:39:48 +02:00
android.md
build-riscv64-spacemit.md
build-s390x.md
build.md CUDA: fix replacment of bad archs in CMake (#18457) 2025-12-29 17:58:20 +01:00
docker.md
function-calling.md
install.md
llguidance.md
multimodal.md
ops.md metal : add count_equal op (#18314) 2025-12-31 10:39:48 +02:00