Hongqiang WangandLi He ea8085373a opencl: add int8 dp4 dense and MoE prefill optimization for Adreno GPUs (llama/25537) ...

* opencl: add int8 dp4 dense and moe GEMM

* opencl: refactor

---------

Co-authored-by: Li He <lih@qti.qualcomm.com>

2026-07-30 16:31:37 +03:00

..

ggml-blas

…

ggml-cann

…

ggml-cpu

ggml : use ggml_vqtbl1q_u8 for 32-bit compat (#0)

2026-07-10 13:06:42 +03:00

ggml-cuda

ggml : process data in smaller chunks in CUDA ggml_top_k() and ggml_argsort() to reduce temporary buffers memory usage (llama/24776)

2026-07-30 16:31:36 +03:00

ggml-et

ggml-et: Initial ET backend (llama/24179)

2026-07-30 16:31:36 +03:00

ggml-hexagon

hexagon: improve ARGSORT performance for small tensors (llama/25512)

2026-07-30 16:31:36 +03:00

ggml-hip

ggml-hip: enable -funsafe-math-optimizations (llama/24668)

2026-07-10 13:06:42 +03:00

ggml-metal

metal : add CONV_2D_DW (depthwise convolution) support (llama/21565)

2026-07-10 13:06:42 +03:00

ggml-musa

…

ggml-opencl

opencl: add int8 dp4 dense and MoE prefill optimization for Adreno GPUs (llama/25537)

2026-07-30 16:31:37 +03:00

ggml-openvino

Added a cudaMemcpy2DAsync fast path to ggml_cuda_cpy (llama/25057)

2026-07-10 13:06:42 +03:00

ggml-rpc

…

ggml-sycl

fix unsupported UT cases of CONT & CPY (llama/25231)

2026-07-10 13:06:42 +03:00

ggml-virtgpu

…

ggml-vulkan

vulkan: disable FA mask_opt on GCN to improve performance (llama/24362)

2026-07-10 13:06:42 +03:00

ggml-webgpu

ggml-webgpu: tune subgroup split (d_split) in flash_attn_vec (llama/25418)

2026-07-10 13:06:42 +03:00

ggml-zdnn

…

ggml-zendnn

…

CMakeLists.txt

ggml-et: Initial ET backend (llama/24179)

2026-07-30 16:31:36 +03:00

ggml-alloc.c

…

ggml-backend-dl.cpp

…

ggml-backend-dl.h

…

ggml-backend-impl.h

…

ggml-backend-meta.cpp

abort if we see a multi buffer (llama/25276)

2026-07-10 13:06:42 +03:00

ggml-backend-reg.cpp

ggml-et: Initial ET backend (llama/24179)

2026-07-30 16:31:36 +03:00

ggml-backend.cpp

Revert "sched : reintroduce less synchronizations during split compute (#20793)" (llama/25138)

2026-07-10 13:06:42 +03:00

ggml-common.h

Add Q2_0 quantization: type definition and CPU backend (llama/24448)

2026-07-10 13:06:42 +03:00

ggml-impl.h

…

ggml-opt.cpp

…

ggml-quants.c

Add Q2_0 quantization: type definition and CPU backend (llama/24448)

2026-07-10 13:06:42 +03:00

ggml-quants.h

Add Q2_0 quantization: type definition and CPU backend (llama/24448)

2026-07-10 13:06:42 +03:00

ggml-threading.cpp

…

ggml-threading.h

…

ggml.c

ggml : add support for CPU f16->f16 GGML_OP_SET_ROWS (llama/25344)

2026-07-10 13:06:42 +03:00

ggml.cpp

…

gguf.cpp

…