Merge branch 'main' of github.com:containers/ramalama into docs

README.md

@@ -90,7 +90,7 @@ curl -fsSL https://raw.githubusercontent.com/containers/ramalama/s/install.sh |
 | Apple Silicon GPU (macOS)          | :white_check_mark: |
 | Apple Silicon GPU (podman-machine) | :x: |
 | Nvidia GPU (cuda)                  | :x: |
-| AMD GPU (rocm)                     | :x: |
+| AMD GPU (rocm)                     | :white_check_mark: |
 
 ## COMMANDS
 

container-images/cpuonly/Containerfile

@@ -3,7 +3,7 @@ FROM registry.access.redhat.com/ubi9/ubi:9.4-1214.1726694543
 # renovate: datasource=github-releases depName=huggingface/huggingface_hub extractVersion=^v(?<version>.*)
 ARG HUGGINGFACE_HUB_VERSION=0.25.2
 # renovate: datasource=github-releases depName=containers/omlmd extractVersion=^v(?<version>.*)
-ARG OMLMD_VERSION=0.1.5
+ARG OMLMD_VERSION=0.1.6
 # renovate: datasource=github-releases depName=tqdm/tqdm extractVersion=^v(?<version>.*)
 ARG TQDM_VERSION=4.66.5
 ARG LLAMA_CPP_SHA=70392f1f81470607ba3afef04aa56c9f65587664

container-images/cuda/Containerfile

@@ -1,9 +1,9 @@
-FROM docker.io/nvidia/cuda:12.6.1-devel-ubi9
+FROM docker.io/nvidia/cuda:12.6.2-devel-ubi9
 
 # renovate: datasource=github-releases depName=huggingface/huggingface_hub extractVersion=^v(?<version>.*)
 ARG HUGGINGFACE_HUB_VERSION=0.25.2
 # renovate: datasource=github-releases depName=containers/omlmd extractVersion=^v(?<version>.*)
-ARG OMLMD_VERSION=0.1.5
+ARG OMLMD_VERSION=0.1.6
 ARG LLAMA_CPP_SHA=70392f1f81470607ba3afef04aa56c9f65587664
 # renovate: datasource=git-refs depName=ggerganov/whisper.cpp packageName=https://github.com/ggerganov/whisper.cpp gitRef=master versioning=loose type=digest
 ARG WHISPER_CPP_SHA=06a1da9daff94c1bf1b1d38950628264fe443f76

container-images/rocm/Containerfile

@@ -14,24 +14,21 @@ RUN rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-Official
 
 RUN dnf install -y rocm-dev hipblas-devel rocblas-devel && \
     dnf clean all && \
-    rm -rf /var/cache/*dnf*
-
-RUN git clone https://github.com/ggerganov/llama.cpp && \
+    git clone https://github.com/ggerganov/llama.cpp && \
     cd llama.cpp && \
     git reset --hard ${LLAMA_CPP_SHA} && \
     cmake -B build -DCMAKE_INSTALL_PREFIX:PATH=/usr -DGGML_CCACHE=0 \
       -DGGML_HIPBLAS=1 && \
     cmake --build build --config Release -j $(nproc) && \
     cmake --install build && \
     cd / && \
-    rm -rf llama.cpp
-
-RUN git clone https://github.com/ggerganov/whisper.cpp.git && \
+    git clone https://github.com/ggerganov/whisper.cpp.git && \
     cd whisper.cpp && \
     git reset --hard ${WHISPER_CPP_SHA} && \
     make -j $(nproc) GGML_HIPBLAS=1 && \
     mv main /usr/bin/whisper-main && \
     mv server /usr/bin/whisper-server && \
     cd / && \
-    rm -rf whisper.cpp
+    rm -rf /var/cache/*dnf* /opt/rocm-*/lib/llvm \
+      /opt/rocm-*/lib/rocblas/library/*gfx9* llama.cpp whisper.cpp
 

container_build.sh

@@ -21,24 +21,30 @@ add_build_platform() {
   conman_build+=("-t" "quay.io/ramalama/$image_name" ".")
 }
 
+rm_container_image() {
+  if [ "$image_name" == "cuda" ] || [ "$image_name" == "rocm" ]; then
+    "$conman_bin" rmi -f "$image_name" || true
+  fi
+}
+
 build() {
   cd "$1"
   local image_name
   image_name=$(echo "$1" | sed "s#container-images/##g")
-  if [ "$image_name" != "rocm" ]; then # todo: skip, trim rocm image, too large
-    local conman_build=("${conman[@]}")
-    if [ "$#" -lt 2 ]; then
-      add_build_platform
-      "${conman_build[@]}" 2>&1 | tee container_build.log
-    elif [ "$2" = "-d" ]; then
-      add_build_platform
-      echo "${conman_build[@]}"
-    elif [ "$2" = "push" ]; then
-      "${conman[@]}" push "quay.io/ramalama/$image_name"
-    else
-      add_build_platform
-      "${conman_build[@]}" 2>&1 | tee container_build.log
-    fi
+  local conman_build=("${conman[@]}")
+  if [ "$#" -lt 2 ]; then
+    add_build_platform
+    "${conman_build[@]}" 2>&1 | tee container_build.log
+    rm_container_image
+  elif [ "$2" = "-d" ]; then
+    add_build_platform
+    echo "${conman_build[@]}"
+  elif [ "$2" = "push" ]; then
+    "${conman[@]}" push "quay.io/ramalama/$image_name"
+  else
+    add_build_platform
+    "${conman_build[@]}" 2>&1 | tee container_build.log
+    rm_container_image
   fi
 
   cd - > /dev/null

ramalama/cli.py

@@ -263,7 +263,7 @@ def _list_containers(args):
     if conman == "":
         raise IndexError("no container manager (Podman, Docker) found")
 
-    conman_args = [conman, "ps", "-a", "--filter", "label=RAMALAMA container"]
+    conman_args = [conman, "ps", "-a", "--filter", "label=RAMALAMA"]
     if args.noheading:
         conman_args += ["--noheading"]
     if hasattr(args, "notrunc") and args.notrunc:
@@ -568,7 +568,7 @@ def run_container(args):
         "--rm",
         "-i",
         "--label",
-        "RAMALAMA container",
+        "RAMALAMA",
         "--security-opt=label=disable",
         "-e",
         "RAMALAMA_TRANSPORT",

test/system/030-run.bats

@@ -8,7 +8,7 @@ load helpers
     model=m_$(safename)
     image=m_$(safename)
 
-    verify_begin="podman run --rm -i --label \"RAMALAMA container\" --security-opt=label=disable -e RAMALAMA_TRANSPORT --name"
+    verify_begin="podman run --rm -i --label RAMALAMA --security-opt=label=disable -e RAMALAMA_TRANSPORT --name"
 
     run_ramalama --dryrun run ${model}
     is "$output" "${verify_begin} ramalama_.*" "dryrun correct"

test/system/040-serve.bats

@@ -2,7 +2,7 @@
 
 load helpers
 
-verify_begin="podman run --rm -i --label \"RAMALAMA container\" --security-opt=label=disable -e RAMALAMA_TRANSPORT --name"
+verify_begin="podman run --rm -i --label RAMALAMA --security-opt=label=disable -e RAMALAMA_TRANSPORT --name"
 
 @test "ramalama --dryrun serve basic output" {
     skip_if_nocontainer