[zhao-stack]

What this PR does / why we need it?

Upgrade the release dependency from vLLM v0.30.0 to v0.31.0 while keeping the independently supported verified-main target at c4df37d. Rebase on latest Ascend main c2382c6, preserving the CPP profiling changes from #18622. The file mapping below describes the diff for 23c8b15 against c2382c6.

The two targets have different APIs despite both being supported: retain version-dependent forwarding for profiling randomization, PP batch lengths, Mamba allocation, EPLB setup and encoder-only prefill admission. Remove obsolete 0.30 branches only where both targets now share the contract.

Contract	Verified main c4df37d	Release v0.31.0 db9527a
Profiling	Parent calls omit randomize_inputs	Forward the flag through V1/V2/310P and MC2
PP batch state	No max_seq_len_np	Preserve the field in AscendInputBatch
Mamba allocation	Parent has no prefill_end	Forward checkpoint boundary
EPLB setup	Update existing state/communicator	Construct state from model and mapping
Non-chunked prefill	Keep encoder-only budget exception	Keep release budget rejection

Changes organized by downstream file

Upstream links identify the reason or contract being adapted. For existing main behavior that becomes common to both targets, the change is removal of the old release branch, not a new port of the complete upstream implementation. File links are pinned to this PR head; the complete fixed diff shows the exact edits.

Dependency markers, images and documentation (13 files)

Downstream file	Upstream change / contract	Ascend adaptation
.github/vllm-release-tag.commit	Release target advances to v0.31.0; verified main remains pinned independently.	Advance the release marker to v0.31.0; leave the verified-main marker at c4df37d.
Dockerfile	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
Dockerfile.310p	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
Dockerfile.310p.openEuler	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
Dockerfile.a3	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
Dockerfile.a3.openEuler	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
Dockerfile.a5	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
Dockerfile.a5.openEuler	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
Dockerfile.openEuler	Release target advances to v0.31.0; verified main remains pinned independently.	Set the release VLLM_TAG default to v0.31.0; existing built images are not changed by this default.
README.md	Release target advances to v0.31.0; verified main remains pinned independently.	Update the release CI support version to 0.31.0.
README.zh.md	Release target advances to v0.31.0; verified main remains pinned independently.	Update the Chinese release CI support version to 0.31.0.
docs/source/tutorials/models/DeepSeek-V4.1-Flash.md	Release target advances to v0.31.0; verified main remains pinned independently.	Use separate release and verified-main marker commands for the two build targets.
vllm_ascend/ops/triton/docs/v2/spec_decode/prepare_dflash_inputs.md	vLLM #57632 passes precompute_context_kv into DFlash FULL-graph capture; wrapped draft models must expose the underlying callback.	Update the documented DFlash ABI version to 0.31.0; no new kernel behavior is introduced by this document change.

Model runners, graphs and platform integration (14 files)

Downstream file	Upstream change / contract	Ascend adaptation
vllm_ascend/_310p/model_runner_310p.py	vLLM #58411 adds randomize_inputs for memory profiling to avoid all-zero MoE input concentration. Present in release; absent from the pinned-main parent calls.	Forward the profiling randomization flag through the 310P V1 override to the NPU V1 runner.
vllm_ascend/_310p/worker/v2/model_runner.py	vLLM #58411 adds randomize_inputs for memory profiling to avoid all-zero MoE input concentration. Present in release; absent from the pinned-main parent calls. vLLM #58400 adds prefill_runs_as_decode_np and decode_graph_eligible for single-token prompt tails with existing context; has_prefill alone no longer determines FULL-decode eligibility. vLLM #56456 adds trailing num_active_loras to prepare_inputs, keeping dispatch and fast-prefill graph/padding selection consistent.	Forward release-only randomization; accept num_active_loras in the common prepare signature, use the shared PCP/input-batch contract and compute complete prompt-tail/decode-graph metadata in CPU input preparation.
vllm_ascend/_310p/worker/v2/sampler.py	vLLM #56497 introduces V2 LogitsContext and custom logits-processor lifecycle; Sampler.add_request no longer takes prompt_len.	Adopt add_request(req_idx, sampling_params) and remove the legacy prompt-length argument; retain the 310P CPU generator/sampling implementation.
vllm_ascend/attention/attention_v1.py	vLLM #58400 adds prefill_runs_as_decode_np and decode_graph_eligible for single-token prompt tails with existing context; has_prefill alone no longer determines FULL-decode eligibility.	Remove the 0.30 condition around V2 speculative-padding handling while retaining actual prefill/decode metadata semantics.
vllm_ascend/patch/worker/patch_v2/patch_model_runner.py	vLLM #56723 adds dcp_sharded to distinguish sharded attention KV from replicated Mamba/GDN state when target DCP and DCP=1 draft coexist. vLLM #45635 moves routed-expert output to AuxOutputConfig/the AuxOutput connector; vLLM #57980 removes the legacy runner initialization keyword.	Read dcp_sharded from cache specs and forward it to block tables; initialize the AuxOutput connector from aux_output_config.enabled; drop the legacy attn_cg_support keyword.
vllm_ascend/platform.py	vLLM #45635 moves routed-expert output to AuxOutputConfig/the AuxOutput connector; vLLM #57980 removes the legacy runner initialization keyword.	Apply the existing V1 AuxOutput configuration guard to both targets, removing the old 0.30 early return.
vllm_ascend/worker/model_runner_v1.py	vLLM #58411 adds randomize_inputs for memory profiling to avoid all-zero MoE input concentration. Present in release; absent from the pinned-main parent calls.	Accept randomize_inputs in profiling/dummy execution and enter maybe_randomize_inputs inside forward context when requested; preserve the Ascend execution path.
vllm_ascend/worker/v2/aclgraph_utils.py	vLLM #57980 replaces the legacy PCP helper with descriptor-based partition_batch, restores auxiliary hidden states with sampling outputs, and renames the DCP preparation helper. vLLM #57632 passes precompute_context_kv into DFlash FULL-graph capture; wrapped draft models must expose the underlying callback.	Use prepare_dcp_local_seq_lens; delegate model-wrapper attribute lookup to the original model so DFlash context-KV callbacks remain accessible.
vllm_ascend/worker/v2/input_batch.py	vLLM #58275/vLLM #57980 expose max_query_len/is_padding on dummy batches and an explicit query-length bound for autoregressive graph capture.	Always retain prompt-tail and graph-eligibility fields and forward dummy-batch max_query_len/is_padding; remove obsolete 0.30 branches.
vllm_ascend/worker/v2/model_runner.py	vLLM #58411 adds randomize_inputs for memory profiling to avoid all-zero MoE input concentration. Present in release; absent from the pinned-main parent calls. vLLM #58542 adds max_seq_len_np for PP sampled-token broadcast handling. Present in release batch state; absent from pinned main. vLLM #58400 adds prefill_runs_as_decode_np and decode_graph_eligible for single-token prompt tails with existing context; has_prefill alone no longer determines FULL-decode eligibility. vLLM #57980 replaces the legacy PCP helper with descriptor-based partition_batch, restores auxiliary hidden states with sampling outputs, and renames the DCP preparation helper. vLLM #56456 adds trailing num_active_loras to prepare_inputs, keeping dispatch and fast-prefill graph/padding selection consistent. vLLM #54335 routes fixed-token scoring through the same unpadded logits path as prompt logprobs, which is incompatible with Ascend lmhead-TP collective row layouts.	Use release-only kwargs for profiling/execute and MC2 dummy execution; forward release-only max_seq_len_np. After PD request reclassification, recompute has_prefill, decode-graph eligibility and uniform token count. Accept num_active_loras in the common prepare signature without adding FastPrefillHelper behavior. Use descriptor-based PCP partitioning and restore auxiliary states before replicated drafting. Extend the lmhead-TP guard to fixed-token scoring, tolerating workers without token_id_scores.
vllm_ascend/worker/v2/model_states/mamba_hybrid.py	vLLM #58400 requires decode-like prompt tails to use decode-row Mamba metadata for speculative rollback.	Build a local mask is_prefilling_np & ~prefill_runs_as_decode_np so prompt tails are decode rows; do not mutate the batch's original prefill mask.
vllm_ascend/worker/v2/pcp_manager.py	vLLM #57980 replaces the legacy PCP helper with descriptor-based partition_batch, restores auxiliary hidden states with sampling outputs, and renames the DCP preparation helper.	Unify descriptor-based partitioning, including CPP profile capture after #18622, and the (batch, hidden, aux) sampling restore contract; restore and trim auxiliary states together with hidden states to the unpadded global token extent.
vllm_ascend/worker/v2/spec_decode/autoregressive/aclgraph.py	vLLM #58275/vLLM #57980 expose max_query_len/is_padding on dummy batches and an explicit query-length bound for autoregressive graph capture.	Always pass desc.max_query_len or desc.uniform_token_count to autoregressive capture.
vllm_ascend/worker/v2/spec_decode/dflash/aclgraph.py	vLLM #57632 passes precompute_context_kv into DFlash FULL-graph capture; wrapped draft models must expose the underlying callback.	Always receive and forward precompute_context_kv under the Ascend communicator/model-capture wrappers.

KV allocation and block tables (6 files)

Downstream file	Upstream change / contract	Ascend adaptation
vllm_ascend/_310p/worker/v2/block_table.py	vLLM #56723 adds dcp_sharded to distinguish sharded attention KV from replicated Mamba/GDN state when target DCP and DCP=1 draft coexist. vLLM #56734 uses idx_mapping=-1 for dummy draft/padding requests and masks their persistent block-table reads to avoid stale-row KV writes.	Unify the constructor with dcp_sharded; skip negative request indices and retain PAD slots in the CPU mapping path. Keep the existing CP=1 restriction and persistent CPU/device buffers.
vllm_ascend/models/glm5next/kv_cache.py	vLLM #59175 adds prefill_end to Mamba checkpoint allocation, including prompt-end retention. Release has the new signature; pinned main has the old one.	Accept prefill_end for call compatibility; keep one private Kpool tail block per request.
vllm_ascend/ops/triton/v2/block_table/compute_slot_mappings.py	vLLM #56734 uses idx_mapping=-1 for dummy draft/padding requests and masks their persistent block-table reads to avoid stale-row KV writes.	Unify the slot-mapping kernel, mask block-table loads with is_real_req, and emit PAD for dummy rows; retain Ascend block/window and address-calculation behavior.
vllm_ascend/patch/platform/patch_kv_cache_utils.py	The pinned _annotate_eagle_groups contract uses use_trailing_layer_fallback, replacing the DeepSeek-V4-specific fallback keyword; the introducing PR was not independently identified. Fixed contracts: release: kv_cache_utils.py.	Always pass use_trailing_layer_fallback=True; retain existing Ascend cache grouping/annotation behavior.
vllm_ascend/patch/platform/patch_mamba_manager.py	vLLM #59175 adds prefill_end to Mamba checkpoint allocation, including prompt-end retention. Release has the new signature; pinned main has the old one.	Accept prefill_end, forwarding it only to the release parent while retaining the existing Ascend allocation logic.
vllm_ascend/worker/v2/block_table.py	vLLM #56723 adds dcp_sharded to distinguish sharded attention KV from replicated Mamba/GDN state when target DCP and DCP=1 draft coexist.	Use one constructor with optional per-group dcp_sharded validation/forwarding; retain Ascend int32 slot mappings.

Schedulers and profiling (5 files)

Downstream file	Upstream change / contract	Ascend adaptation
vllm_ascend/core/dyntra_lb_scheduler.py	vLLM #59029 gives main an encoder-only exception to the non-chunked prefill budget check; release keeps the budget rejection condition. vLLM #58947 replaces skipped_waiting with kv_holding_waiting and deferred_waiting, prioritizes KV holders, and uses _handle_blocked_waiting_request. vLLM #58947 and vLLM #58259 define the upstream resumable-session queue/handoff lifecycle: stopping a turn can re-enqueue buffered continuation input. Both targets admit against max_num_active_reqs and include parked streaming sessions in occupied slots; related queue/Elastic-EP changes are vLLM #58947 and vLLM #53934. Fixed contracts: release: scheduler.py, release: scheduler.py.	Use KV-holder-first queue selection/re-enqueueing, the common blocked-request helper, deferred bookkeeping and active-slot admission including parked streams. Preserve the main-only encoder exception. Remove secondary stop cleanup so upstream can retain/re-enqueue resumable continuations.
vllm_ascend/core/recompute_scheduler.py	vLLM #59029 gives main an encoder-only exception to the non-chunked prefill budget check; release keeps the budget rejection condition. vLLM #58947 replaces skipped_waiting with kv_holding_waiting and deferred_waiting, prioritizes KV holders, and uses _handle_blocked_waiting_request. Both targets admit against max_num_active_reqs and include parked streaming sessions in occupied slots; related queue/Elastic-EP changes are vLLM #58947 and vLLM #53934. Fixed contracts: release: scheduler.py, release: scheduler.py.	Fold the legacy queues into the common KV-holder/deferred path, count parked streams for active admission, and preserve the version-dependent encoder condition. Remove the redundant remote-KV waiting debug log as a local cleanup.
vllm_ascend/core/scheduler_profiling_chunk.py	vLLM #59029 gives main an encoder-only exception to the non-chunked prefill budget check; release keeps the budget rejection condition. vLLM #58947 replaces skipped_waiting with kv_holding_waiting and deferred_waiting, prioritizes KV holders, and uses _handle_blocked_waiting_request. Both targets admit against max_num_active_reqs and include parked streaming sessions in occupied slots; related queue/Elastic-EP changes are vLLM #58947 and vLLM #53934. Fixed contracts: release: scheduler.py, release: scheduler.py.	Use the common blocked-request/queue lifecycle, active admission including parked streams, and the version-dependent encoder budget condition.
vllm_ascend/patch/platform/patch_balance_schedule.py	vLLM #59029 gives main an encoder-only exception to the non-chunked prefill budget check; release keeps the budget rejection condition. vLLM #58947 replaces skipped_waiting with kv_holding_waiting and deferred_waiting, prioritizes KV holders, and uses _handle_blocked_waiting_request.	Fold six obsolete 0.30 queue branches and add the release/main encoder budget condition. Document the shared if not request_queue drained-queue guard. Retain baseline balancing, running-only DP snapshots/admission accounting and connector-snapshot cleanup; complete upstream schedule synchronization is outside this patch.
vllm_ascend/profiling_config.py	vLLM #58947 replaces skipped_waiting with kv_holding_waiting and deferred_waiting, prioritizes KV holders, and uses _handle_blocked_waiting_request.	Retarget the profiling wrapper from the old blocked-status promotion helper to _handle_blocked_waiting_request.

EPLB state and routing (2 files)

Downstream file	Upstream change / contract	Ascend adaptation
vllm_ascend/distributed/eplb/eplb_state.py	vLLM #53934 changes main's EPLB setup to update existing state and reuse its communicator. Release still creates state from (model, model_config, mapping).	After the parent's update_mapping, refresh Ascend replica routing tables; consistently resolve an unspecified NPU device index to the current device.
vllm_ascend/worker/v2/eplb.py	vLLM #53934 changes main's EPLB setup to update existing state and reuse its communicator. Release still creates state from (model, model_config, mapping).	Create AscendEplbState on release with the current policy/optional expert arguments; call the main parent to update the existing state and communicator.

Models and MoE operators (6 files)

Downstream file	Upstream change / contract	Ascend adaptation
vllm_ascend/models/kimi_k3.py	vLLM #50592/vLLM #50593 use prefix_delta and the three-value AttnRes layer contract instead of the old prepared-input/deferred-MLP contract.	Remove old layer/model contracts, propagate prefix_delta and return the three-value AttnRes state; preserve Ascend fused operations, DSpark aux selection and SP/PP transport/output boundaries.
vllm_ascend/ops/fused_moe/gate_linear.py	vLLM #58234 unifies MoE gate layers under GateLinear with the current linear constructor and replaces Step3.5's dedicated FP32 gate class.	Use the current constructor parameters (skip_bias_add, quant_config, return_bias) and retain BF16-to-FP32 gate/fallback semantics. CRLF-to-LF normalization contributes to the large line diff.
vllm_ascend/ops/fused_moe/router/grouped_topk_router.py	vLLM #58880 removes routed_scaling_factor from routing-method classification, allowing MiniMax classification with non-unit scaling.	Stop passing routed_scaling_factor to classification; downstream routing-weight scaling is retained.
vllm_ascend/patch/__init__.py	The release target is 0.31.0; vLLM #56994 is context for the retained GLM reasoning patch comment, not a new model capability in this PR.	Update only the supported-version comment for the retained GLM reasoning patch.
vllm_ascend/patch/worker/patch_step3p5.py	vLLM #58234 unifies MoE gate layers under GateLinear with the current linear constructor and replaces Step3.5's dedicated FP32 gate class.	Always use GateLinear with FP32 parameter/output types instead of the legacy FP32ReplicatedLinear branch.

Token scoring and Triton helpers (2 files)

Downstream file	Upstream change / contract	Ascend adaptation
vllm_ascend/patch/worker/patch_triton.py	vLLM #58651 uses triton.next_power_of_2 in packed-QK RoPE constexpr evaluation; the Ascend replacement must also satisfy the installed Triton JIT/constexpr-function contract.	Use a constexpr-aware next_power_of_2 replacement; register with triton.constexpr_function when available, otherwise unwrap tl.constexpr.value in the fallback.
vllm_ascend/worker/v2/sample/logprob.py	vLLM #54335 adds fixed-token prompt scoring and broadcasts candidate IDs with expand; the token-ID row stride can be zero or non-contiguous.	Pass token_ids.stride(0) into the Triton kernel and use it for row addressing; support broadcast stride zero and strided candidate IDs.

BalanceScheduler scope: this PR folds the old queue branches and adapts encoder-only admission. Its baseline DP balancing, running-only snapshot/accounting and connector-snapshot cleanup remain; complete scheduler-body synchronization and a strict AST equivalence requirement are deferred. The baseline drift guard may skip and does not prove complete upstream equivalence.

Does this PR introduce any user-facing change?

The default release dependency becomes vLLM v0.31.0; the verified-main pin remains unchanged. Fixed-token prompt scoring handles broadcast/strided IDs correctly and reports an explicit error when used with unsupported lmhead TP. Resumable Dyntra continuations retain upstream stop/queue behavior, and PD prompt-tail graph/Mamba metadata follow the supported upstream contracts.

How was this patch tested?

Test changes organized by file

The table describes test intent and fixture changes; it does not claim that every changed test was selected or executed on NPU hardware.

Unit, PR E2E and nightly test mapping (37 files)

Downstream file	Upstream change / contract	Ascend adaptation
tests/e2e/nightly/single_node/ops/singlecard_ops/triton/test_compute_token_logprobs.py	vLLM #54335	Add NPU numerical comparisons for broadcast stride=0 and sliced stride=8 token IDs; outside the PR CI selector.
tests/e2e/nightly/single_node/ops/singlecard_ops/triton/test_num_nans.py	Release-version cleanup; no new upstream behavior.	Update the default version label to 0.31.0; no algorithm change.
tests/e2e/pull_request/four_card/test_kimi_k3.py	vLLM #54222	Check equality between P-worker remote_prefill_cached_tokens and the response cache count; zero hits are valid for a new prompt.
tests/e2e/pull_request/one_card/test_model_runner_v1_with_device.py	vLLM #45635, vLLM #57980	Use the common AuxOutput configuration object in V1 guard fixtures.
tests/e2e/pull_request/one_card/test_thinking_budget.py	vLLM #56497	Construct LogitsContext with expanded mapping, input IDs, positions and sequence bounds; call apply(logits, ctx).
tests/ut/_310p/test_model_runner_v2_310p.py	vLLM #56456, vLLM #56497	Use the current sampler request signature and pass the added prepare-input LoRA argument in 310P fixtures.
tests/ut/attention/test_mla_cp.py	vLLM #58149 makes ModelState.max_model_len a model-config-backed property; vLLM #57913 moves multimodal capability to ModelConfig; vLLM #56723 updates DCP state/metadata contracts.	Mock get_dcp_world_size_and_rank and remove legacy DCP group fixture branching.
tests/ut/core/test_circular_buffer.py	Pinned cache allocation requires replay boundaries (vLLM #58368); DeepSeek-V4.1 architecture metadata and Engram configuration (vLLM #56214) use their current configuration contracts. Individual configuration migrations are linked to fixed source where not precisely attributed. Fixed contracts: release: single_type_kv_cache_manager.py, release: model_arch.py, release: engram.py.	Pass replay_boundaries=[tokens-1] to the current cache-block API.
tests/ut/core/test_dyntra_lb_recompute_scheduler.py	vLLM #58947	Use the shared KV-holder/deferred queue interfaces in Dyntra/recompute fixtures.
tests/ut/core/test_dyntra_lb_scheduler.py	vLLM #58259, vLLM #58947	Regress synchronous/asynchronous resumable continuation handling for running, paused and preempted requests, buffered/unbuffered input, and final cleanup; update queue seams.
tests/ut/core/test_profiling_chunk.py	vLLM #53934, vLLM #58947	Update queue fixtures/blocked-request helper and active-request capacity field.
tests/ut/core/test_recompute_scheduler.py	vLLM #58947	Remove legacy queue branching and use the shared blocked-request seam.
tests/ut/core/test_scheduler_connector_block_state.py	vLLM #58947, vLLM #59029	Supply the baseline balance_queue fixture; connector-snapshot cleanup itself predates this change.
tests/ut/distributed/eplb/test_state.py	vLLM #53934	Verify mapping updates refresh Ascend routing tables.
tests/ut/distributed/kv_transfer/kv_p2p/test_remote_prefill_lifecycle.py	vLLM #58947	Use the shared queue lifecycle rather than version-specific skipped-waiting fixtures.
tests/ut/kv_offload/test_preempt_offload_connector.py	Pinned cache allocation requires replay boundaries (vLLM #58368); DeepSeek-V4.1 architecture metadata and Engram configuration (vLLM #56214) use their current configuration contracts. Individual configuration migrations are linked to fixed source where not precisely attributed. Fixed contracts: release: single_type_kv_cache_manager.py, release: model_arch.py, release: engram.py.	Always provide prefix replay and request token-count fields in the offload fixture.
tests/ut/models/test_deepseek_v41_release_config.py	Pinned cache allocation requires replay boundaries (vLLM #58368); DeepSeek-V4.1 architecture metadata and Engram configuration (vLLM #56214) use their current configuration contracts. Individual configuration migrations are linked to fixed source where not precisely attributed. Fixed contracts: release: single_type_kv_cache_manager.py, release: model_arch.py, release: engram.py.	Assert obsolete flattened multimodal-prefix flags are absent after architecture conversion.
tests/ut/models/test_glm5next_kv_cache.py	vLLM #59175	Pass the new allocation argument while checking the private-tail block policy.
tests/ut/models/test_kimi_k3_adapter.py	vLLM #50592, vLLM #50593	Update layer signatures/three-value returns, auxiliary output expectations and AST-extracted fixtures.
tests/ut/models/test_kimi_k3_pp_sp.py	vLLM #50592, vLLM #50593	Use the common Kimi AttnRes contract in PP/SP checks.
tests/ut/patch/platform/test_patch_balance_schedule.py	vLLM #58947, vLLM #59029	Use common queue fixtures; add eight encoder/chunked/prompt-budget cases and empty/fully deferred queue regressions with/without KV blocks. Correct the structural guard assertion to if not request_queue. Keep the baseline AST drift check, which can skip.
tests/ut/patch/platform/test_patch_engram_config.py	Pinned cache allocation requires replay boundaries (vLLM #58368); DeepSeek-V4.1 architecture metadata and Engram configuration (vLLM #56214) use their current configuration contracts. Individual configuration migrations are linked to fixed source where not precisely attributed. Fixed contracts: release: single_type_kv_cache_manager.py, release: model_arch.py, release: engram.py.	Expect shared-memory default resolution for this supported fixture configuration; no universal shared-memory default is claimed.
tests/ut/patch/platform/test_prefix_cache_cp_patches.py	vLLM #59175	Supply an explicit prefill_end allocation boundary.
tests/ut/spec_decode/test_eagle_proposer.py	vLLM #58149 makes ModelState.max_model_len a model-config-backed property; vLLM #57913 moves multimodal capability to ModelConfig; vLLM #56723 updates DCP state/metadata contracts.	Use the current model-config multimodal capability interface and remove obsolete skips/branches.
tests/ut/spec_decode/test_mla_dspark.py	vLLM #57632	Assert DFlash context-KV callback forwarding and the shared capture signature.
tests/ut/test_minimax_m3_kv_block_size.py	Release-version cleanup; no new upstream behavior.	Remove only an unused version-helper import.
tests/ut/test_platform.py	vLLM #45635, vLLM #57980	Use the current AuxOutput configuration guard for both target versions.
tests/ut/worker/test_attn_utils_v2.py	vLLM #58149 makes ModelState.max_model_len a model-config-backed property; vLLM #57913 moves multimodal capability to ModelConfig; vLLM #56723 updates DCP state/metadata contracts.	Initialize model_state.model_config instead of assigning the read-only length property.
tests/ut/worker/test_dcp_speculator_metadata_v2.py	vLLM #58149 makes ModelState.max_model_len a model-config-backed property; vLLM #57913 moves multimodal capability to ModelConfig; vLLM #56723 updates DCP state/metadata contracts.	Assert prefill metadata over the padded request extent.
tests/ut/worker/test_model_runner_v2.py	vLLM #58411 adds randomize_inputs for memory profiling to avoid all-zero MoE input concentration. Present in release; absent from the pinned-main parent calls. vLLM #58542 adds max_seq_len_np for PP sampled-token broadcast handling. Present in release batch state; absent from pinned main. vLLM #58400 adds prefill_runs_as_decode_np and decode_graph_eligible for single-token prompt tails with existing context; has_prefill alone no longer determines FULL-decode eligibility. vLLM #57980 replaces the legacy PCP helper with descriptor-based partition_batch, restores auxiliary hidden states with sampling outputs, and renames the DCP preparation helper. vLLM #56456 adds trailing num_active_loras to prepare_inputs, keeping dispatch and fast-prefill graph/padding selection consistent. vLLM #54335 routes fixed-token scoring through the same unpadded logits path as prompt logprobs, which is incompatible with Ascend lmhead-TP collective row layouts. vLLM #58149 makes ModelState.max_model_len a model-config-backed property; vLLM #57913 moves multimodal capability to ModelConfig; vLLM #56723 updates DCP state/metadata contracts.	Cover release/main kwargs and PP fields, PD graph reclassification, PCP auxiliary restoration and lmhead scoring guards; update model-state fixtures.
tests/ut/worker/test_model_runner_v2_encoder_decoder.py	vLLM #58149 makes ModelState.max_model_len a model-config-backed property; vLLM #57913 moves multimodal capability to ModelConfig; vLLM #56723 updates DCP state/metadata contracts.	Update model-state length fixtures through model configuration.
tests/ut/worker/test_model_runner_v2_mamba.py	vLLM #58400	Regress prompt-tail masks for FULL/NONE graph modes and zero/three draft tokens; verify the original mask is unchanged.
tests/ut/worker/test_pcp_manager_v2.py	vLLM #45635, vLLM #57980	Assert descriptor-based partitioning and hidden/aux restore layout; update AuxOutput fixture configuration.
tests/ut/worker/test_v2_dummy_padding.py	vLLM #58400 adds prefill_runs_as_decode_np and decode_graph_eligible for single-token prompt tails with existing context; has_prefill alone no longer determines FULL-decode eligibility. vLLM #58275/vLLM #57980 expose max_query_len/is_padding on dummy batches and an explicit query-length bound for autoregressive graph capture. vLLM #58149 makes ModelState.max_model_len a model-config-backed property; vLLM #57913 moves multimodal capability to ModelConfig; vLLM #56723 updates DCP state/metadata contracts.	Use the model-config-backed state/dummy descriptor contract and retain graph-eligibility metadata.
tests/ut/worker/v2/test_aclgraph.py	vLLM #57980, vLLM #58275	Update capture descriptors and dummy-batch keyword expectations.
tests/ut/worker/v2/test_eplb_controller.py	vLLM #53934	Cover release state construction versus main in-place state/communicator reuse.
tests/ut/worker/v2/test_eplb_load_collection_phase.py	vLLM #53934	Match each target's EPLB setup signature in load-collection fixtures.

Local A3 nightly validation (2026-10-15; frozen head 444e962)

These four model runs used 444e962b239c3a8fc56b9255d601818e6e028b9a + vLLM 0.31.0 (db9527a), which is also the restored PR head; the deferred connector hardening is not included. Prefill 138 + Decode 136, 16 NPUs per node; original image, weights, datasets, launch and benchmark parameters were retained, with only the second-node address mapped from 139 to 136. All four cases exited pytest 0 on both nodes (one passed JUnit case per node). OTT is output token/s; TPOT/TTFT are ms.

Case	Performance requests	OTT / required minimum	TPOT / TTFT	AIME2025 accuracy	Result
Kimi-K2.6 PD	60/60	355.1593 / 337.024075	20.5 / 3268.2	90%, within 93.33±10	Performance + accuracy passed
DeepSeek-V4 Flash PD-prefix	Accuracy only	N/A	N/A	90%, within 90±10	Accuracy passed
GLM-5.1 W8A8C8 128K	256/256	811.7645 / 282.105876	39.8 / 32665.6	No accuracy item	Performance passed
MiniMax-M2.7 PD-prefix	160/160	711.0518 / 698.497	34.8 / 17438.2	90%, within 93.33±10	Performance + accuracy passed

All performance requests above, and the valid controls below, were checked for actual 1024-token outputs, not just success flags.

Conditional controls: historical throughput differences triggered only the affected performance cases against the same PR base 6d1dd70, without this PR, + vLLM 0.30.0 (ced6857). Kimi: 355.1593 vs 359.8514 OTT (-1.30%); the earlier 7.27% loss was not reproduced in this comparison. TPOT was 20.5 vs 19.7ms (+0.8ms/+4.06%), and average request latency +3.48%; a single pair does not establish latency equivalence or a stable regression. GLM: 811.7645 vs 813.3506 OTT (-0.20%), TPOT 39.8 vs 39.4ms. Kimi's original random request pacing affects aggregate OTT. No alternating reruns or unaffected-model 0.30 controls were added.

Archived evidence: main2main-nightly-20261013/pr18229-update-444e962-20261015/ (FINAL_REPORT.md, per-node logs/JUnit, raw AISBench outputs, token-count audit, controls and diagnostic evidence). Both nodes were released after testing.

Deferred follow-up: multi-group receive-failure reporting (not included)

The request-level receive-failure hardening change has been removed from this upgrade PR; the branch is restored to the nightly-tested head 444e962. The original four nightlies already passed on this head without that fix, so it is not required for these normal-path workloads to pass.

The fault-path issue remains real: if a multi-group receive actually fails, the existing connector can report flat block IDs and trigger the vLLM 0.31 scheduler guard, terminating Decode EngineCore. This is handled separately in #18823, retaining ordinary non-HMA block-level recovery semantics. The upgrade PR does not claim fault isolation is fixed.

Development fault-injection results belong to the deferred connector implementation, not this restored head. No new NPU run was performed when restoring the exact previously tested code.

Historical validation for head 23c8b15

• New E2E workflow and Docs link check were triggered after the rebase. Results are pending; previous-head CI does not validate this rebased head.
• A targeted comment command requests test_deepseek_v4_mtp_full_decode_only against both supported upstream targets. The command workflow is queued; its actual test results remain pending.
• Ruff check and format check passed for all 72 changed Python files; Python compilation and diff checks passed. Source-extracted execution of both real CPP routing regressions passed against the required partition signatures of the exact main/release refs. This validates dispatch contracts with lightweight dependency stubs and does not replace full pytest or NPU execution.
• Existing source-extracted scheduler regressions passed on both version lanes: eight budget scenarios, two queue paths and empty/fully deferred queue regressions, including negative controls.
• Local full pytest/Torch/NPU and bash format.sh ci remain unavailable in this Windows environment. They require CI validation.
• Previous head 9330488 passed pre-commit, Docs and both CPU UT lanes (main 10782 passed / 114 skipped; release 10784 passed / 112 skipped). Its initial NPU matrix passed 41/42 shards; the remaining main A3 shard failed in unchanged upstream teardown with psutil.NoSuchProcess, before the MTP golden acceptance assertion. The consequent gate failure and old-head rerun are historical evidence, not a result for the current head.

Coverage limits and historical evidence

The PR CI selector scans tests/ut and tests/e2e/pull_request; the new broadcast/strided token-logprob numerical tests under tests/e2e/nightly are outside that selection. A5, standalone upstream-worker and coverage-upload jobs were skipped in the inspected run. Skips are not coverage. No dedicated packed-QK RoPE or V1 profiling-randomization standalone UT is added by this final diff.

Previous head 15f8a3d passed E2E and Docs, including 42 NPU jobs. Its PR-recorded CPU counts were 10574 passed / 113 skipped for verified main and 10576 passed / 111 skipped for release, with no failures. Those counts belong to that historical head and are not asserted as current-head counts.

• vLLM main: vllm-project/vllm@c4df37d

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request performs an incremental upgrade of the vLLM dependency to v0.31.0. It incorporates necessary adaptations to Ascend-specific scheduling, worker, and model runner logic to maintain compatibility with upstream architectural changes, particularly regarding scheduler queue management and input batching. The PR also cleans up legacy compatibility code that is no longer required with the new version pin.

Highlights

• vLLM Version Upgrade: Upgraded vLLM main and release pins to v0.31.0, requiring updates to scheduling and worker logic.
• Scheduler Adaptations: Updated recompute, profiling, and Dyntra LB schedulers to align with upstream changes to queue management and blocked request handling.
• Worker and Model Runner Updates: Refactored input batching, profiling passthrough, and KV cache coordination to support v0.31.0 contracts.
• Legacy Path Retirement: Removed obsolete compatibility paths for v0.29/v0.30, simplifying the codebase for the new version.

New Features

🧠 You can now enable Memory (public preview) to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize the Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counterproductive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[github-actions]

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

• A PR should do only one thing, smaller PRs enable faster reviews.
• Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
• Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

[gemini-code-assist]

Code Review

Suggested PR Title:\n\nmarkdown\n[Ops][Misc] Align vllm_ascend with vLLM v0.31.0 API changes\n\n\nSuggested PR Summary:\n\nmarkdown\n### What this PR does / why we need it?\nThis pull request updates the `vllm_ascend` repository to maintain compatibility with the upstream vLLM v0.31.0 release. It adapts various modules, including attention backends, schedulers, model runners, and the Kimi K3 model, to match upstream API refactorings (such as `LogitsContext`, `AuxOutputConfig`, `EplbState` attribute renames, and `PCPManager` changes). Additionally, it refactors `AscendGateLinear` to support quantization and bias configurations, and fixes a potential out-of-bounds memory access in the Triton slot mapping kernel when handling dummy requests.\n\nFeedback: A potential issue was identified in the Triton slot mapping kernel where a negative pointer offset is calculated for dummy requests (`idx_mapping == -1`). Clamping the index to a valid row is recommended to prevent memory access faults on NPU architectures.\n\n### Does this PR introduce _any_ user-facing change?\nNo, these are internal alignment and compatibility updates for the Ascend NPU backend.\n\n### How was this patch tested?\nThe changes are covered by updating the existing unit tests and end-to-end tests to match the new vLLM v0.31.0 contracts.\n

[zhao-stack]

/rerun
[Bot]: rerun completed.

Rerun (failed jobs only):

• E2E

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[zhao-stack]

/rerun
[Bot]: rerun completed.

Rerun (failed jobs only):

• E2E

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[zhao-stack]

/rerun
[Bot]: rerun completed.

Rerun (failed jobs only):

• E2E

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[zhao-stack]

/rerun
[Bot]: rerun completed.

Rerun (failed jobs only):

• E2E

[AuroraEmiya]

LGTM