build.go: introduce a friendlier way to build Ollama (#3548)

This commit introduces a more friendly way to build Ollama dependencies
and the binary without abusing `go generate` and removing the
unnecessary extra steps it brings with it.

This script also provides nicer feedback to the user about what is
happening during the build process.

At the end, it prints a helpful message to the user about what to do
next (e.g. run the new local Ollama).

.github/workflows/release.yaml

@@ -95,7 +95,8 @@ jobs:
           cd $env:GITHUB_WORKSPACE
           $env:CMAKE_SYSTEM_VERSION="10.0.22621.0"
           $env:PATH="$gopath;$env:PATH"
-          go generate -x ./...
+          
+          $env:GOARCH = ""; go run build.go -f -d -target=${{ matrix.arch }}
         name: go generate
       - uses: actions/upload-artifact@v4
         with:

.github/workflows/test.yaml

@@ -1,5 +1,16 @@
 name: test
 
+concurrency:
+  # For PRs, later CI runs preempt previous ones. e.g. a force push on a PR
+  # cancels running CI jobs and starts all new ones.
+  #
+  # For non-PR pushes, concurrency.group needs to be unique for every distinct
+  # CI run we want to have happen. Use run_id, which in practice means all
+  # non-PR CI runs will be allowed to run without preempting each other.
+  group: ${{ github.workflow }}-$${{ github.pull_request.number || github.run_id }}
+  cancel-in-progress: true
+
+
 on:
   pull_request:
     paths:
@@ -62,10 +73,12 @@ jobs:
           $env:CMAKE_SYSTEM_VERSION="10.0.22621.0"
           $env:PATH="$gopath;$gccpath;$env:PATH"
           echo $env:PATH
-          go generate -x ./...
+
+          $env:GOARCH = ""; go run build.go -f -d -target=${{ matrix.arch }}
         if: ${{ startsWith(matrix.os, 'windows-') }}
         name: 'Windows Go Generate'
-      - run: go generate -x ./...
+      - run: |
+          GOARCH= go run build.go -f -d -target=${{ matrix.arch }}
         if: ${{ ! startsWith(matrix.os, 'windows-') }}
         name: 'Unix Go Generate'
       - uses: actions/upload-artifact@v4
@@ -98,7 +111,7 @@ jobs:
       - run: go get ./...
       - run: |
           git config --global --add safe.directory /__w/ollama/ollama
-          go generate -x ./...
+          GOARCH= go run build.go -f -d -target=${{ matrix.arch }}
         env:
           OLLAMA_SKIP_CPU_GENERATE: '1'
       - uses: actions/upload-artifact@v4
@@ -129,7 +142,7 @@ jobs:
       - run: go get ./...
       - run: |
           git config --global --add safe.directory /__w/ollama/ollama
-          go generate -x ./...
+          GOARCH= go run build.go -f -d -target=${{ matrix.arch }}
         env:
           OLLAMA_SKIP_CPU_GENERATE: '1'
       - uses: actions/upload-artifact@v4
@@ -168,8 +181,9 @@ jobs:
           $env:PATH="$gopath;$env:PATH"
           $env:OLLAMA_SKIP_CPU_GENERATE="1"
           $env:HIP_PATH=$(Resolve-Path 'C:\Program Files\AMD\ROCm\*\bin\clang.exe' | split-path | split-path)
-          go generate -x ./...
-        name: go generate
+
+          $env:GOARCH = ""; go run build.go -f -d -target=${{ matrix.arch }}
+        name: go run build.go
         env:
           OLLAMA_SKIP_CPU_GENERATE: '1'
       # TODO - do we need any artifacts?
@@ -202,7 +216,7 @@ jobs:
       - name: 'Verify CUDA'
         run: nvcc -V
       - run: go get ./...
-      - name: go generate
+      - name: go run build.go
         run: |
           $gopath=(get-command go).source | split-path -parent
           $cudabin=(get-command nvcc).source | split-path
@@ -211,7 +225,8 @@ jobs:
           $env:CMAKE_SYSTEM_VERSION="10.0.22621.0"
           $env:PATH="$gopath;$cudabin;$env:PATH"
           $env:OLLAMA_SKIP_CPU_GENERATE="1"
-          go generate -x ./...
+          
+          $env:GOARCH = ""; go run build.go -f -d -target=${{ matrix.arch }}
         env:
           OLLAMA_SKIP_CPU_GENERATE: '1'
       # TODO - do we need any artifacts?
@@ -285,6 +300,12 @@ jobs:
         with:
           go-version-file: go.mod
           cache: true
+      - run: |
+          GOARCH= go run build.go -f -d -target=${{ matrix.arch }}
+        if: ${{ ! startsWith(matrix.os, 'windows-') }}
+      - run: |
+          $env:GOARCH = ""; go run build.go -f -d -target=${{ matrix.arch }}
+        if: ${{ startsWith(matrix.os, 'windows-') }}
       - run: go get
       - run: |
           case ${{ matrix.arch }} in
@@ -305,9 +326,8 @@ jobs:
           touch llm/build/windows/$ARCH/stub/bin/ollama_llama_server
         if: ${{ startsWith(matrix.os, 'windows-') }}
         shell: bash
-      - run: go generate ./...
-      - run: go build
-      - run: go test -v ./...
+      - run: |
+          go test -v ./...
       - uses: actions/upload-artifact@v4
         with:
           name: ${{ matrix.os }}-binaries