Conversation
Contributor
Author
|
Empirical evidence (gfx1101, ROCm/HIP 7.15 container Routes/gating:
End-to-end turn: UI-shaped streamed request (
CI note: the |
ghazni101
force-pushed
the
feat/serve-chat-ui-beta
branch
from
October 2, 2026 23:16
b2d8573 to
2d9042d
Compare
added 3 commits
October 3, 2026 08:39
`(b >> 2j) & 0x3 <= 3` is always true; clippy's deny-level bad_bit_mask stopped `cargo clippy --all-targets` at hipfire-quantize. Pin the block geometry instead.
The no-GPU CI runner has no /opt/rocm/core-10.0, so peacemaker-ir's codec gates panicked in the lib-test step and 14 hipfire-isa integration tests in the GPU-free step. Follow the existing table-gate convention: skip with a log line when llvm-mc is absent; the gates still run where it exists.
clippy's deny-by-default mut_from_ref stopped `cargo clippy --all-targets` at hipfire-xdna once hipfire-quantize compiled. Bo is a cloneable handle to one shared mapping, so &mut self would not prevent aliasing; exclusivity is the documented caller contract. Same fix as a1ac7f4 on feat/qwen38-flash-next.
`(b >> 2j) & 0x3 <= 3` is always true; clippy's deny-level bad_bit_mask stopped `cargo clippy --all-targets` at hipfire-quantize. Pin the block geometry instead.
The no-GPU CI runner has no /opt/rocm/core-10.0, so peacemaker-ir's codec gates panicked in the lib-test step and 15 hipfire-isa integration tests in the GPU-free step. Follow the existing table-gate convention: skip with a log line when llvm-mc is absent; the gates still run where it exists.
clippy's deny-by-default mut_from_ref stopped `cargo clippy --all-targets` at hipfire-xdna once hipfire-quantize compiled. Bo is a cloneable handle to one shared mapping, so &mut self would not prevent aliasing; exclusivity is the documented caller contract. Same fix as a1ac7f4 on feat/qwen38-flash-next.
…ad lifecycle a31e438 (land/041t) grew crates/hipfire-daemon/src/main.rs to 5392 lines without moving the ceiling, so beta's own gates job fails. The 15 lines are no_model_loaded_suffix (an admitted single-device/pp load failure now says the prior model is gone) and its unit test. RATCHET-RAISE: daemon_lines 5377 -> 5392, traded for the Flash-Next unload-first load lifecycle (a31e438) that fixes the FN<->VMM swap refusal and names the missing model on admitted-load failure
…E_UI) Port of feat/serve-chat-ui-v040 onto the 0.4.x beta line (the v040 branch is authored against master; serve/complete.rs, serve/mod.rs and serve/http.rs have since been refactored on beta, so the Rust hunks were re-applied against the beta shapes rather than rebased). What ships: - Zero-dependency chat frontend: five static assets embedded via include_str! (index/app/db/render/style), served by a new serve/ui.rs matcher that shares the listener and the existing /v1/chat/completions path. IndexedDB history, branching, markdown/highlighting, effort picker, live TTFT/tok-s strip, mobile layout, ember palette. - Enablement in precedence order: --ui, --open (implies --ui, launches the browser), serve.ui = true, HIPFIRE_SERVE_UI=1. Off is byte-for-byte the old behavior: / and /ui both 404. Non-loopback binds warn on stderr; detached children get --ui. - Security headers on every UI route: strict CSP (default-src 'none', frame-ancestors 'none', form-action 'none'), X-Frame-Options: DENY, nosniff, no-cache — the page is an unauthenticated control surface and is not frameable. - /health and /v1/models publish the resident model's reasoning contract and effort rungs (from meta.loaded — no runtime lock), and capabilities gains chat_ui. - Carries the two gateway fixes the UI surfaced: mid-generation daemon failures without attested rollback poison the session so the next request cold-resets (poisons_session_state), and completion_timings no longer reports wall tok/s as decode tok/s. (slots.rs unchanged: beta already clamps max_tokens_fit via hipfire_generate::common::fit_max_tokens_if.) Verified on gfx1101 against a fresh --ui --no-prewarm serve: / -> 302 /ui; /ui 200 with the full header set; all four assets 200 with correct content types; non-UI paths still 404; /health reports chat_ui: true and the reasoning fields. cargo test -p hipfire-cli 312/312; crate maps regenerated.
The module-level use sat unused on beta; the ported test now resolves through it instead of shadowing with its own import (clears the unused-import advisory).
The UI held a single global in-flight guard (state.stream), serializing every chat through one request at a time even when serve.multi_slot runs the daemon's concurrent slot engine. Streams are now keyed by conversation id (state.streams: Map), with curStream()/anyStreaming() helpers. - send/regenerate/edit/tool-reply/variant-switch block only on the same conversation's stream; other chats generate in parallel - stop button and Esc abort the open conversation's stream only - deleteConv/clearAll/bc del+reload abort the matching streams - sidebar conv-streaming dot shows for every in-flight chat - per-conversation debounced saves (a shared timer dropped the second stream's writes) - view updates/scroll only for the open conversation; streaming view is nulled when switching away and rebound on return - doc title shows ● Generating… while any stream runs hidden - busy toast rewritten: 'keep working in another chat' Verified against a stubbed /v1/chat/completions SSE backend driving the real page in headless Chromium over CDP: two conversations streamed concurrently (streams.size=2, both lens growing), same-chat send refused mid-stream, stop aborted only the open chat, deleteConv aborted its stream, sidebar dots tracked in-flight convs.
…lot thinking-on) Every thinking-on (or thinking-auto) request on the multi-slot route failed closed with 'reasoning prefix mismatch: expected OpenThink got ClosedThink'. The route derived the continuation/reasoning framing as OpenThink / ClosedThink / Plain from `has_think`, which is the *tokenizer's* inventory of a `<think>` special token — not the prompt. Standard Qwen3-family templates emit no opener when thinking is enabled and let the model generate `<think>` itself (only enable_thinking=false emits a closed `<think>\n\n</think>\n\n` block), so a thinking-on render ends in a bare `assistant\n`: started_in_think=false and has_think=true fabricated ClosedThink, which then failed the equality check against the request-derived expected OpenThink. The Jinja render owns the generation suffix, so the framing is read back from what the template actually emitted: new render_assistant_prefix() (sibling of render_tail_opens_think, tail-only for the same reason) maps a closed `</think>` tail to ClosedThink, an open `<think>` tail to OpenThink, and anything else to Plain. The VL and ChatFrame builders frame the assistant turn themselves from expected_prefix and keep returning it. The fail-closed equality check is removed: it compared two independent derivations of the same quantity, and the render is the authoritative one. started_in_think (what emit_generation_start latches) is now simply prefix == OpenThink. Reproduced against the MiMo-V2.6-Distill-Qwen-9B container on this branch: 'hi' with no thinking field, enable_thinking=true both returned the mismatch error; enable_thinking=false worked. Also present on upstream/beta and upstream/master (slots.rs:1168/1172) — not a regression from this branch. Test: render_tail_think_tests::rendered_framing_is_read_from_the_suffix.
…hinking Second gate on the same thinking-on path, reached once the prefix check was fixed: 'think_mode mismatch with enable_thinking' (typed internal, so a 500). The route derived think_mode from the prompt framing (started_in_think) and then failed closed unless it agreed with the request's enable_thinking. Those are different quantities: enable_thinking is a Jinja *input*, and a thinking-on Qwen3-family template frames no opener — the model emits <think> itself, so started_in_think is false while enable_thinking is true. The parser must start outside the reasoning span and open it on that token, which is what the framing-derived value already says. Drop the equality check and keep think_mode = framing. Also drop the now-stale 'think' case from the post-start error-release seam list (the remaining three still cover the internal class).
- Deleted-chat toast no longer covers the composer: #toasts anchors below the topbar (top: calc(var(--topbar-h) + 12px)) instead of the bottom edge, and the enter/leave keyframes drop from above to match. - Each sidebar row gains a visible delete button beside the three-dot menu, wired to the same deleteConv path (undo toast included). It shares the three-dot button's reveal: hidden until row hover/active/focus, always visible on touch/narrow. Hover tints it with the error colour. - Drop the topbar chat-title button (redundant with the sidebar row): the Rename action now targets the sidebar row, opening the sidebar first when it is collapsed. Removes .conv-title/.conv-title-input and the responsive hide. The delete entry stays in the three-dot menu as before.
ghazni101
force-pushed
the
feat/serve-chat-ui-beta
branch
from
October 3, 2026 14:28
31cea59 to
96388a7
Compare
…beta # Conflicts: # crates/hipfire-daemon/map.md # docs/env-vars.md
…erve-chat-ui-beta
…ilure suffix helper out of main.rs land/041t + land/041u grew daemon main.rs to 5392 against the 5377 ceiling, leaving beta's own gates job red for every branch based on its head. Raising the ceiling needs the maintainer-only 'ratchet-raise' label, so pay the debt instead: no_model_loaded_suffix and its test move to request_guards (the module whose charter is exactly 'small load/generate message helpers kept out of main.rs'), semantics untouched — main.rs lands at 5372. Also refreshes the stale crate maps (engine/generate/quantize/cli/config/daemon).
… (main.rs is back under it)
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.
Summary
A zero-dependency chat frontend embedded in the serve binary, off by default:
hipfire serve --ui(or--open,serve.ui = true,HIPFIRE_SERVE_UI=1) serves it at/uion the same listener and the same/v1/chat/completionspath every other client uses. Chat history lives in the browser's IndexedDB — serve stores nothing. Off is byte-for-byte the old behavior (/and/ui404).Which surface(s) does this touch?
crates/hipfire-cli/src/serve(gateway routes, session-reset flag),crates/hipfire-config(serve.uikey)hipfire-cliflags, config key, docsslots.rs— unchanged: beta already clampsmax_tokens_fitviahipfire_generate::common::fit_max_tokens_ifDetails
assets/ui/(index/app/db/render/style), embedded viainclude_str!; no new crates, no JS toolchain. History sidebar with pin/search/export/import, message branching (edit/regenerate forks), markdown + syntax highlighting + TeX-unicode math, reasoning disclosure, effort picker, image attach gated on the model's vision capability, live TTFT/tok-s strip, mobile layout, theme switcher.default-src 'none'; script-src 'self'; frame-ancestors 'none'; form-action 'none',X-Frame-Options: DENY, nosniff, no-cache. Unit tests assert the CSP value (clickjacking class closed, not just header presence)./healthand/v1/modelsarchitecturepublish the resident model's reasoning contract and effort rungs (served frommeta.loaded, no runtime lock);capabilitiesgainschat_ui.rolled_back: falsenow poison session state so the next request cold-resets (PR serve: cold-reset the next request after a mid-generation failure #806);completion_timingsno longer reports walltok_sasdecode_tok_s(PR serve: completion timings no longer report wall tok/s as decode tok/s #807). Both are needed for the UI to be correct; keeping them here too so this branch is self-contained — drop either hunk if the master fix lands first.feat/serve-chat-ui-v040(master-based) onto beta; serve/complete/mod/http were refactored since, so hunks were re-applied against beta's shapes (slot-side clamp already exists on beta and was dropped).Test plan
cargo build --release -p hipfire-cli -p hipfire-daemoncleancargo test -p hipfire-cli— 312/312 (incl. new ui route/header, reasoning-facts, poison-classification,serve_ui_url/host_is_loopbackunit tests)scripts/check-crate-maps.py --check,scripts/check-env-docs.py,scripts/check-lifecycle.py,scripts/leanup-ratchets.shgreen; crate maps regenerated--ui --no-prewarm):/302 →/ui200, all assets 200 with correct content types + full header set, non-UI paths 404,/healthreportschat_ui: trueand the reasoning fields. Earlier iteration verified headless at 320/360/390/768px (no overflow/overlap) and a full streamed chat turn against qwen3.5-4b.mq4 multi-slot.serve_harness.py battery— the gateway changes are additive (new routes + flag) but the poison-fix touches session lifecycle; recommend a harness pass on a flagship artifact. Daemon binary unchanged.Hardware validation request (optional)
{ "routes": [ {"mode": "battery", "tag": "qwen3.8:27b-mq4-xts"} ], "claim": "serve gateway unchanged for non-UI requests; session cold-reset path behaves on poisoned multi-slot sessions" }