upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-23 11:10:55 -05:00
Code Issues Packages Projects Releases Wiki Activity

sycl : Make GGML_SYCL_F16=ON the default (#23996)

Browse Source 
* Add -cl-fp32-correctly-rounded-divide-sqrt to F16=ON builds

Signed-off-by: Todd Malsbary <todd.malsbary@intel.com>

* Make GGML_SYCL_F16=ON the default

Signed-off-by: Todd Malsbary <todd.malsbary@intel.com>

* Leave F32 the default

F16 remains explictly set for example and Dockerfile builds.

Signed-off-by: Todd Malsbary <todd.malsbary@intel.com>

* Revert changes to examples/sycl/build scripts

Signed-off-by: Todd Malsbary <todd.malsbary@intel.com>

---------

Signed-off-by: Todd Malsbary <todd.malsbary@intel.com>
This commit is contained in:
Todd Malsbary
2026-06-15 22:34:02 -07:00
committed by GitHub
parent ad39ccaa19
commit 4196b477da
2 changed files with 15 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
.devops/intel.Dockerfile
View File

@@ -7,7 +7,7 @@ ARG APP_REVISION=N/A
FROM docker.io/intel/deep-learning-essentials:$ONEAPI_VERSION AS build
ARG GGML_SYCL_F16=OFF
ARG GGML_SYCL_F16=ON
ARG LEVEL_ZERO_VERSION=1.28.2
ARG LEVEL_ZERO_UBUNTU_VERSION=u24.04
RUN apt-get update && \
@@ -24,7 +24,8 @@ COPY . .
RUN if [ "${GGML_SYCL_F16}" = "ON" ]; then \
echo "GGML_SYCL_F16 is set" \
&& export OPT_SYCL_F16="-DGGML_SYCL_F16=ON"; \
&& export OPT_SYCL_F16="-DGGML_SYCL_F16=ON" \
&& export SYCL_PROGRAM_COMPILE_OPTIONS="-cl-fp32-correctly-rounded-divide-sqrt"; \
fi && \
echo "Building with dynamic libs" && \
cmake -B build -DGGML_NATIVE=OFF -DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DGGML_BACKEND_DL=ON -DGGML_CPU_ALL_VARIANTS=ON -DLLAMA_BUILD_TESTS=OFF ${OPT_SYCL_F16} && \