Skip to content

Commit 043579b

Browse files
committed
rebase
1 parent 38add61 commit 043579b

File tree

4 files changed

+19
-19
lines changed

4 files changed

+19
-19
lines changed

go.mod

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -29,7 +29,7 @@ require (
2929
sigs.k8s.io/gateway-api-inference-extension v0.0.0-20260128235548-fd30cb97714a
3030
)
3131

32-
replace sigs.k8s.io/gateway-api-inference-extension => github.com/RishabhSaini/gateway-api-inference-extension v0.0.0-20260202150317-4d55e2564b01
32+
replace sigs.k8s.io/gateway-api-inference-extension => github.com/RishabhSaini/gateway-api-inference-extension v0.0.0-20260202163711-ec0927b44aa2
3333

3434
require (
3535
cel.dev/expr v0.24.0 // indirect

go.sum

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -16,8 +16,8 @@ github.com/AzureAD/microsoft-authentication-library-for-go v1.6.0 h1:XRzhVemXdgv
1616
github.com/AzureAD/microsoft-authentication-library-for-go v1.6.0/go.mod h1:HKpQxkWaGLJ+D/5H8QRpyQXA1eKjxkFlOMwck5+33Jk=
1717
github.com/Masterminds/semver/v3 v3.4.0 h1:Zog+i5UMtVoCU8oKka5P7i9q9HgrJeGzI9SA1Xbatp0=
1818
github.com/Masterminds/semver/v3 v3.4.0/go.mod h1:4V+yj/TJE1HU9XfppCwVMZq3I84lprf4nC11bSS5beM=
19-
github.com/RishabhSaini/gateway-api-inference-extension v0.0.0-20260202150317-4d55e2564b01 h1:TWmpkx/DH6LasXPCGYkbyIugalQuiEvcZTvw6qWb7v8=
20-
github.com/RishabhSaini/gateway-api-inference-extension v0.0.0-20260202150317-4d55e2564b01/go.mod h1:lvMpB9a+Lk+xBi5Pk6teUG+NqA16WR8nRpmBNFJbflU=
19+
github.com/RishabhSaini/gateway-api-inference-extension v0.0.0-20260202163711-ec0927b44aa2 h1:lumgh/rLgHbYx77JECUKZ7dsxHm0ZcTs7JEXUgo3nHI=
20+
github.com/RishabhSaini/gateway-api-inference-extension v0.0.0-20260202163711-ec0927b44aa2/go.mod h1:lvMpB9a+Lk+xBi5Pk6teUG+NqA16WR8nRpmBNFJbflU=
2121
github.com/alecthomas/units v0.0.0-20240927000941-0f3dac36c52b h1:mimo19zliBX/vSQ6PWWSL9lK8qwHozUj03+zLoEB8O0=
2222
github.com/alecthomas/units v0.0.0-20240927000941-0f3dac36c52b/go.mod h1:fvzegU4vN3H1qMT+8wDmzjAcDONcgo2/SZ/TyfdUOFs=
2323
github.com/alicebob/miniredis/v2 v2.35.0 h1:QwLphYqCEAo1eu1TqPRN2jgVMPBweeQcR21jeqDCONI=

pkg/plugins/register.go

Lines changed: 13 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -10,17 +10,17 @@ import (
1010

1111
// RegisterAllPlugins registers the factory functions of all plugins in this repository.
1212
func RegisterAllPlugins() {
13-
plugins.Register(filter.ByLabelType, filter.ByLabelFactory)
14-
plugins.Register(filter.ByLabelSelectorType, filter.ByLabelSelectorFactory)
15-
plugins.Register(filter.DecodeRoleType, filter.DecodeRoleFactory)
16-
plugins.Register(filter.PrefillRoleType, filter.PrefillRoleFactory)
17-
plugins.Register(prerequest.PrefillHeaderHandlerType, prerequest.PrefillHeaderHandlerFactory)
18-
plugins.Register(profile.DataParallelProfileHandlerType, profile.DataParallelProfileHandlerFactory)
19-
plugins.Register(profile.PdProfileHandlerType, profile.PdProfileHandlerFactory)
20-
plugins.Register(scorer.PrecisePrefixCachePluginType, scorer.PrecisePrefixCachePluginFactory)
21-
plugins.Register(scorer.LoadAwareType, scorer.LoadAwareFactory)
22-
plugins.Register(scorer.SessionAffinityType, scorer.SessionAffinityFactory)
23-
plugins.Register(scorer.ActiveRequestType, scorer.ActiveRequestFactory)
24-
plugins.Register(scorer.NoHitLRUType, scorer.NoHitLRUFactory)
25-
plugins.Register(scorer.PDSLOAwareScorerType, scorer.PDSLOAwareScorerFactory)
13+
plugin.Register(filter.ByLabelType, filter.ByLabelFactory)
14+
plugin.Register(filter.ByLabelSelectorType, filter.ByLabelSelectorFactory)
15+
plugin.Register(filter.DecodeRoleType, filter.DecodeRoleFactory)
16+
plugin.Register(filter.PrefillRoleType, filter.PrefillRoleFactory)
17+
plugin.Register(prerequest.PrefillHeaderHandlerType, prerequest.PrefillHeaderHandlerFactory)
18+
plugin.Register(profile.DataParallelProfileHandlerType, profile.DataParallelProfileHandlerFactory)
19+
plugin.Register(profile.PdProfileHandlerType, profile.PdProfileHandlerFactory)
20+
plugin.Register(scorer.PrecisePrefixCachePluginType, scorer.PrecisePrefixCachePluginFactory)
21+
plugin.Register(scorer.LoadAwareType, scorer.LoadAwareFactory)
22+
plugin.Register(scorer.SessionAffinityType, scorer.SessionAffinityFactory)
23+
plugin.Register(scorer.ActiveRequestType, scorer.ActiveRequestFactory)
24+
plugin.Register(scorer.NoHitLRUType, scorer.NoHitLRUFactory)
25+
plugin.Register(scorer.PDSLOAwareScorerType, scorer.PDSLOAwareScorerFactory)
2626
}

pkg/plugins/scorer/pd_slo_aware_router_hooks.go

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -107,12 +107,12 @@ func (p *PDSLOAwareRouter) recordPrefillTrainingData(
107107

108108
// P/D-specific: Extract prefill pod from the "prefill" profile
109109
prefillResult, exists := schedulingResult.ProfileResults["prefill"]
110-
if !exists || prefillResult == nil || len(prefillResult.TargetPods) == 0 {
110+
if !exists || prefillResult == nil || len(prefillResult.TargetEndpoints) == 0 {
111111
logger.V(logutil.DEBUG).Info("No prefill pod in scheduling result, skipping prefill training")
112112
return
113113
}
114114

115-
prefillPod := prefillResult.TargetPods[0]
115+
prefillPod := prefillResult.TargetEndpoints[0]
116116

117117
// Get metrics for the prefill pod
118118
lastSeenMetrics, err := p.PredictedLatency.GetLastSeenMetricsForRequest(request)
@@ -163,7 +163,7 @@ func (p *PDSLOAwareRouter) recordPrefillTrainingData(
163163
logger.V(logutil.DEBUG).Error(err, "Failed to record prefill training data")
164164
} else {
165165
logger.V(logutil.DEBUG).Info("Recorded prefill training data",
166-
"pod", prefillPod.GetPod().String(),
166+
"pod", prefillPod.GetMetadata().String(),
167167
"ttft_ms", actualPrefillTTFT,
168168
"pod_type", "prefill")
169169
}

0 commit comments

Comments
 (0)