]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 511636d)
Fix Docker ROCM builds, use AMDGPU_TARGETS instead of GPU_TARGETS (#9641)
authorserhii-nakon <redacted>
Mon, 30 Sep 2024 18:57:12 +0000 (21:57 +0300)
committerGitHub <redacted>
Mon, 30 Sep 2024 18:57:12 +0000 (20:57 +0200)
commit6f1d9d71f4c568778a7637ff6582e6f6ba5fb9d3
treedda1d9aeffef5ce34aff2db41be4dedb003f7232tree
parent511636df0c90826b4dd1fc21ff260c19d69a3b5dcommit | diff
Fix Docker ROCM builds, use AMDGPU_TARGETS instead of GPU_TARGETS (#9641)

* Fix Docker ROCM builds, use AMDGPU_TARGETS instead of GPU_TARGETS

* Set ROCM_DOCKER_ARCH as string due it incorrectly build and cause OOM exit code
.devops/full-rocm.Dockerfile diff | blob | history
.devops/llama-cli-rocm.Dockerfile diff | blob | history
.devops/llama-server-rocm.Dockerfile diff | blob | history
.github/workflows/build.yml diff | blob | history
Packaging of ggml-org/llama.cpp