ExternalVendorCode/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2024-12-18 20:27:57 +00:00
Code Issues Actions 12 Packages Projects Releases Wiki Activity

fix(cuda): downgrade to 12.0 to increase compatibility range (#2994)

Browse Source 
* fix(cuda): downgrade to 12.0 to increase compatibility range

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* improve messaging

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2024-07-23 23:35:31 +02:00 committed by GitHub
parent 1c96e0b79e
commit a9757fb057
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
6 changed files with 11 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
.github/workflows/image-pr.yml vendored
View File

@ -47,7 +47,7 @@ jobs:
# makeflags: "--jobs=3 --output-sync=target"
- build-type: 'cublas'
cuda-major-version: "12"
cuda-minor-version: "4"
cuda-minor-version: "0"
platforms: 'linux/amd64'
tag-latest: 'false'
tag-suffix: '-cublas-cuda12-ffmpeg'
@ -120,7 +120,7 @@ jobs:
# makeflags: "--jobs=3 --output-sync=target"
# - build-type: 'cublas'
# cuda-major-version: "12"
# cuda-minor-version: "4"
# cuda-minor-version: "0"
# platforms: 'linux/amd64'
# tag-latest: 'false'
# tag-suffix: '-cublas-cuda12-ffmpeg-core'

8
.github/workflows/image.yml vendored
View File

@ -75,7 +75,7 @@ jobs:
makeflags: "--jobs=3 --output-sync=target"
- build-type: 'cublas'
cuda-major-version: "12"
cuda-minor-version: "4"
cuda-minor-version: "0"
platforms: 'linux/amd64'
tag-latest: 'false'
tag-suffix: '-cublas-cuda12'
@ -100,7 +100,7 @@ jobs:
makeflags: "--jobs=3 --output-sync=target"
- build-type: 'cublas'
cuda-major-version: "12"
cuda-minor-version: "4"
cuda-minor-version: "0"
platforms: 'linux/amd64'
tag-latest: 'auto'
tag-suffix: '-cublas-cuda12-ffmpeg'
@ -285,7 +285,7 @@ jobs:
makeflags: "--jobs=4 --output-sync=target"
- build-type: 'cublas'
cuda-major-version: "12"
cuda-minor-version: "4"
cuda-minor-version: "0"
platforms: 'linux/amd64'
tag-latest: 'false'
tag-suffix: '-cublas-cuda12-core'
@ -307,7 +307,7 @@ jobs:
makeflags: "--jobs=4 --output-sync=target"
- build-type: 'cublas'
cuda-major-version: "12"
cuda-minor-version: "4"
cuda-minor-version: "0"
platforms: 'linux/amd64'
tag-latest: 'false'
tag-suffix: '-cublas-cuda12-ffmpeg-core'

1
.github/workflows/release.yaml vendored
View File

@ -31,7 +31,6 @@ jobs:
with:
go-version: '1.21.x'
cache: false
- name: Dependencies
run: |
sudo apt-get update

2
Dockerfile
View File

@ -99,7 +99,7 @@ FROM requirements-${IMAGE_TYPE} AS requirements-drivers
ARG BUILD_TYPE
ARG CUDA_MAJOR_VERSION=12
ARG CUDA_MINOR_VERSION=4
ARG CUDA_MINOR_VERSION=0
ENV BUILD_TYPE=${BUILD_TYPE}

2
Makefile
View File

@ -480,7 +480,7 @@ prepare-e2e:
mkdir -p $(TEST_DIR)
cp -rfv $(abspath ./tests/e2e-fixtures)/gpu.yaml $(TEST_DIR)/gpu.yaml
test -e $(TEST_DIR)/ggllm-test-model.bin || wget -q https://huggingface.co/TheBloke/CodeLlama-7B-Instruct-GGUF/resolve/main/codellama-7b-instruct.Q2_K.gguf -O $(TEST_DIR)/ggllm-test-model.bin
docker build --build-arg GRPC_BACKENDS="$(GRPC_BACKENDS)" --build-arg IMAGE_TYPE=core --build-arg BUILD_TYPE=$(BUILD_TYPE) --build-arg CUDA_MAJOR_VERSION=12 --build-arg CUDA_MINOR_VERSION=4 --build-arg FFMPEG=true -t localai-tests .
docker build --build-arg GRPC_BACKENDS="$(GRPC_BACKENDS)" --build-arg IMAGE_TYPE=core --build-arg BUILD_TYPE=$(BUILD_TYPE) --build-arg CUDA_MAJOR_VERSION=12 --build-arg CUDA_MINOR_VERSION=0 --build-arg FFMPEG=true -t localai-tests .
run-e2e-image:
ls -liah $(abspath ./tests/e2e-fixtures)

6
pkg/model/initializers.go
View File

@ -212,7 +212,7 @@ func selectGRPCProcess(backend, assetDir string, f16 bool) string {
grpcProcess = p
foundCUDA = true
} else {
log.Info().Msgf("GPU device found but no CUDA backend present")
log.Debug().Msgf("Nvidia GPU device found, no embedded CUDA variant found. You can ignore this message if you are using container with CUDA support")
}
}
if strings.Contains(gpu.String(), "amd") {
@ -222,7 +222,7 @@ func selectGRPCProcess(backend, assetDir string, f16 bool) string {
grpcProcess = p
foundAMDGPU = true
} else {
log.Info().Msgf("GPU device found but no HIPBLAS backend present")
log.Debug().Msgf("AMD GPU device found, no embedded HIPBLAS variant found. You can ignore this message if you are using container with HIPBLAS support")
}
}
if strings.Contains(gpu.String(), "intel") {
@ -236,7 +236,7 @@ func selectGRPCProcess(backend, assetDir string, f16 bool) string {
grpcProcess = p
foundIntelGPU = true
} else {
log.Info().Msgf("GPU device found but no Intel backend present")
log.Debug().Msgf("Intel GPU device found, no embedded SYCL variant found. You can ignore this message if you are using container with SYCL support")
}
}
}