highperfocused/ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2025-11-11 08:07:20 +01:00
Code Issues Packages Projects Releases Wiki Activity

DRY out the runner lifecycle code (#12540)

Browse Source 
* DRY out the runner lifecycle code

Now that discovery uses the runners as well, this unifies the runner spawning code
into a single place.  This also unifies GPU discovery types with the newer ml.DeviceInfo

* win: make incremental builds better

Place build artifacts in discrete directories so incremental builds don't have to start fresh

* Adjust sort order to consider iGPUs

* handle cpu inference oom scenarios

* review comments
This commit is contained in:
Daniel Hiltgen
2025-10-23 11:20:02 -07:00
committed by GitHub
parent 1c093e97af
commit 3258a89b6e
16 changed files with 720 additions and 924 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
server/routes_generate_test.go
View File

@@ -17,9 +17,9 @@ import (
"github.com/google/go-cmp/cmp"
"github.com/ollama/ollama/api"
"github.com/ollama/ollama/discover"
"github.com/ollama/ollama/fs/ggml"
"github.com/ollama/ollama/llm"
"github.com/ollama/ollama/ml"
)
type mockRunner struct {
@@ -48,8 +48,8 @@ func (mockRunner) Tokenize(_ context.Context, s string) (tokens []int, err error
return
}
func newMockServer(mock *mockRunner) func(discover.GpuInfoList, string, *ggml.GGML, []string, []string, api.Options, int) (llm.LlamaServer, error) {
return func(_ discover.GpuInfoList, _ string, _ *ggml.GGML, _, _ []string, _ api.Options, _ int) (llm.LlamaServer, error) {
func newMockServer(mock *mockRunner) func(ml.SystemInfo, []ml.DeviceInfo, string, *ggml.GGML, []string, []string, api.Options, int) (llm.LlamaServer, error) {
return func(_ ml.SystemInfo, _ []ml.DeviceInfo, _ string, _ *ggml.GGML, _, _ []string, _ api.Options, _ int) (llm.LlamaServer, error) {
return mock, nil
}
}
@@ -157,9 +157,9 @@ func TestGenerateChat(t *testing.T) {
loaded: make(map[string]*runnerRef),
newServerFn: newMockServer(&mock),
getGpuFn: getGpuFn,
getCpuFn: getCpuFn,
getSystemInfoFn: getSystemInfoFn,
waitForRecovery: 250 * time.Millisecond,
loadFn: func(req *LlmRequest, _ *ggml.GGML, _ discover.GpuInfoList, _ bool) bool {
loadFn: func(req *LlmRequest, _ *ggml.GGML, _ ml.SystemInfo, _ []ml.DeviceInfo, _ bool) bool {
// add small delay to simulate loading
time.Sleep(time.Millisecond)
req.successCh <- &runnerRef{
@@ -768,9 +768,9 @@ func TestGenerate(t *testing.T) {
loaded: make(map[string]*runnerRef),
newServerFn: newMockServer(&mock),
getGpuFn: getGpuFn,
getCpuFn: getCpuFn,
getSystemInfoFn: getSystemInfoFn,
waitForRecovery: 250 * time.Millisecond,
loadFn: func(req *LlmRequest, _ *ggml.GGML, _ discover.GpuInfoList, _ bool) bool {
loadFn: func(req *LlmRequest, _ *ggml.GGML, _ ml.SystemInfo, _ []ml.DeviceInfo, _ bool) bool {
// add small delay to simulate loading
time.Sleep(time.Millisecond)
req.successCh <- &runnerRef{
@@ -1193,9 +1193,9 @@ func TestChatWithPromptEndingInThinkTag(t *testing.T) {
loaded: make(map[string]*runnerRef),
newServerFn: newMockServer(mock),
getGpuFn: getGpuFn,
getCpuFn: getCpuFn,
getSystemInfoFn: getSystemInfoFn,
waitForRecovery: 250 * time.Millisecond,
loadFn: func(req *LlmRequest, _ *ggml.GGML, _ discover.GpuInfoList, _ bool) bool {
loadFn: func(req *LlmRequest, _ *ggml.GGML, _ ml.SystemInfo, _ []ml.DeviceInfo, _ bool) bool {
time.Sleep(time.Millisecond)
req.successCh <- &runnerRef{llama: mock}
return false