fix(profiles): withdraw amd-gcn's CPU alt route - no model runs on CPU - #482
Merged
Merged
Conversation
Operator order 2026-09-24. amd-gcn no longer declares alt_backends ["cpu"], so `install render --llama-bin-cpu` refuses on it by name and a re-render cannot bring offload-e4b-cpu / gemma4-e2b-cpu back. The node that carried the family was edited live to match (entries, ldcpu macro and matrix vars removed, alt_backends dropped from its manifest; /fleet/health reports ["vulkan"]). ADR 0054 is amended, not superseded: the declare/render/advertise mechanism is unchanged and stays available for a non-CPU alternate. Tier docs regenerated with cmd/gentiers. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Intent: operator order 2026-09-24 - no model runs on CPU. Remove the CPU seat family from the one tier that declared it, so a re-render cannot bring it back.
What changed
setup/templates/profiles.json:amd-gcndropsalt_backends: ["cpu"]; its notes and the field description say why and when.install render --llama-bin-cpunow refuses on this tier by name (ADR 0054's "flag without the declaration is an error").docs/tiers/amd-gcn.mdregenerated withgo run ./cmd/gentiers(only that page changed).The live node was edited to match before this PR: the two
*-cpuentries, theldcpumacro and their matrix vars removed from its llama-swap config (backed up first),alt_backendsdropped frominstalled.json, fleet node restarted;/fleet/healthnow reportsbackends: ["vulkan"]and no*-cpuid is served.How tested:
go build ./...,go vet ./...,go test ./...all green locally (90 packages ok). The altcpu renderer tests build their params directly, so they still cover the mechanism.Risk: low - one tier-table field plus docs; no code path changes.
🤖 Generated with Claude Code