merge and debug tbo on 0.9.2

vllm/two_batch_overlap/v1/model_input_split_v1.py

+
+from typing import Any, Optional, Union
+import numpy as np
+import torch
+from vllm import envs
+from vllm.distributed.kv_transfer.kv_transfer_state import get_kv_transfer_group, has_kv_transfer_group
+from vllm.distributed.parallel_state import get_pp_group, get_tp_group
+from vllm.forward_context import set_forward_context
+from vllm.sequence import IntermediateTensors
+from vllm.two_batch_overlap.v1.two_batch_overlap_v1 import tbo_model_executable_v1
+from vllm.utils import async_tensor_h2d
+from vllm.v1.attention.backends.utils import CommonAttentionMetadata
+from vllm.v1.core.sched.output import CachedRequestData, SchedulerOutput
+from vllm.v1.outputs import ModelRunnerOutput
+from vllm.v1.spec_decode.metadata import SpecDecodeMetadata
+from vllm.v1.worker.block_table import BlockTable
+
+
+class TBOModelInputSplit():
+    def __init__(self):
+        self.req_ids_left = []
+        self.req_ids_right = []
+        self.req_num_left = 0
+        self.req_num_right = 0
+        self.scheduler_output_left = None
+        self.scheduler_output_right = None
+
+input_split = TBOModelInputSplit()
+
+def split_scheduler_output(runner, scheduler_output:SchedulerOutput):
+    split_tokens = scheduler_output.total_num_scheduled_tokens // 2
+    req_ids = runner.input_batch.req_ids
+    tokens_counter = 0
+    min_idx = -1
+    min_counter = 0
+    for i, id in enumerate(req_ids):
+        tokens_counter += scheduler_output.num_scheduled_tokens[id]
+        diff = abs(tokens_counter - split_tokens)
+        if min_idx == -1 or diff < min_counter:
+            min_idx = i
+            min_counter = diff
+        if tokens_counter > split_tokens or diff == 0:
+            break
+    input_split.req_num_left = min_idx + 1
+    if input_split.req_num_left == len(req_ids):
+        input_split.req_num_left = input_split.req_num_left - 1
+    input_split.req_ids_left = req_ids[:input_split.req_num_left]
+    input_split.req_ids_right = req_ids[input_split.req_num_left:]
+    input_split.req_num_right = len(req_ids) - input_split.req_num_left
+    new_req_data_left = []
+    new_req_data_right = []
+    cached_reqs_left = []
+    cached_reqs_right = []
+    num_scheduled_tokens_left = {}
+    num_scheduled_tokens_right = {}
+    total_num_scheduled_tokens_left = 0
+    total_num_scheduled_tokens_right = 0
+    for new_req in scheduler_output.scheduled_new_reqs:
+        if new_req.req_id in input_split.req_ids_left:
+            new_req_data_left.append(new_req)
+        else:
+            new_req_data_right.append(new_req)
+
+    #print('###scheduler_output.scheduled_cached_reqs', scheduler_output.scheduled_cached_reqs)
+    cached_reqs_left = CachedRequestData.make_empty()
+    cached_reqs_right = CachedRequestData.make_empty()
+    for req_idx, req_id in enumerate(scheduler_output.scheduled_cached_reqs.req_ids):
+        if req_id in input_split.req_ids_left:
+            cached_reqs_left.req_ids.append(req_id)
+            cached_reqs_left.resumed_from_preemption.append(scheduler_output.scheduled_cached_reqs.resumed_from_preemption[req_idx])
+            if len(scheduler_output.scheduled_cached_reqs.new_token_ids) > 0:
+                cached_reqs_left.new_token_ids.append(scheduler_output.scheduled_cached_reqs.new_token_ids[req_idx])
+            cached_reqs_left.new_block_ids.append(scheduler_output.scheduled_cached_reqs.new_block_ids[req_idx])
+            cached_reqs_left.num_computed_tokens.append(scheduler_output.scheduled_cached_reqs.num_computed_tokens[req_idx])
+        else:
+            cached_reqs_right.req_ids.append(req_id)
+            cached_reqs_right.resumed_from_preemption.append(scheduler_output.scheduled_cached_reqs.resumed_from_preemption[req_idx])
+            if len(scheduler_output.scheduled_cached_reqs.new_token_ids) > 0:
+                cached_reqs_right.new_token_ids.append(scheduler_output.scheduled_cached_reqs.new_token_ids[req_idx])
+            cached_reqs_right.new_block_ids.append(scheduler_output.scheduled_cached_reqs.new_block_ids[req_idx])
+            cached_reqs_right.num_computed_tokens.append(scheduler_output.scheduled_cached_reqs.num_computed_tokens[req_idx])
+    for key, value in scheduler_output.num_scheduled_tokens.items():
+        if key in input_split.req_ids_left:
+            num_scheduled_tokens_left[key] = value
+            total_num_scheduled_tokens_left += value
+        else:
+            num_scheduled_tokens_right[key] = value
+            total_num_scheduled_tokens_right += value
+
+
+    input_split.scheduler_output_left = SchedulerOutput(
+        scheduled_new_reqs=new_req_data_left,
+        scheduled_cached_reqs=cached_reqs_left,
+        num_scheduled_tokens=num_scheduled_tokens_left,
+        total_num_scheduled_tokens=total_num_scheduled_tokens_left,
+        scheduled_spec_decode_tokens=scheduler_output.scheduled_spec_decode_tokens,
+        scheduled_encoder_inputs=scheduler_output.scheduled_encoder_inputs, ##unsupport yet
+        num_common_prefix_blocks=scheduler_output.num_common_prefix_blocks,
+        # finished_req_ids is an existing state in the scheduler,
+        # instead of being newly scheduled in this step.
+        # It contains the request IDs that are finished in between
+        # the previous and the current steps.
+        finished_req_ids=scheduler_output.finished_req_ids,
+        free_encoder_input_ids=scheduler_output.free_encoder_input_ids,
+        structured_output_request_ids=scheduler_output.structured_output_request_ids,
+        grammar_bitmask=scheduler_output.grammar_bitmask,
+    )
+    input_split.scheduler_output_right = SchedulerOutput(
+        scheduled_new_reqs=new_req_data_right,
+        scheduled_cached_reqs=cached_reqs_right,
+        num_scheduled_tokens=num_scheduled_tokens_right,
+        total_num_scheduled_tokens=total_num_scheduled_tokens_right,
+        scheduled_spec_decode_tokens=scheduler_output.scheduled_spec_decode_tokens,
+        scheduled_encoder_inputs=scheduler_output.scheduled_encoder_inputs, ##unsupport yet
+        num_common_prefix_blocks=scheduler_output.num_common_prefix_blocks,
+        # finished_req_ids is an existing state in the scheduler,
+        # instead of being newly scheduled in this step.
+        # It contains the request IDs that are finished in between
+        # the previous and the current steps.
+        finished_req_ids=scheduler_output.finished_req_ids,
+        free_encoder_input_ids=scheduler_output.free_encoder_input_ids,
+        structured_output_request_ids=scheduler_output.structured_output_request_ids,
+        grammar_bitmask=scheduler_output.grammar_bitmask,
+    )
+
+
+def prepare_tbo_atten_metadata(
+    runner,
+    scheduler_output: "SchedulerOutput",
+    req_ids,
+    req_offset
+) -> tuple[dict[str, Any], torch.Tensor, Optional[SpecDecodeMetadata]]:
+    total_num_scheduled_tokens = scheduler_output.total_num_scheduled_tokens
+    assert total_num_scheduled_tokens > 0
+    num_reqs = len(req_ids)
+    assert num_reqs > 0
+
+    seq_len_offset = req_offset
+    if req_offset == 0: #left
+        query_start_offset = 0
+    else:
+        query_start_offset = req_offset + 1
+
+    # Get the number of scheduled tokens for each request.
+    tokens = [scheduler_output.num_scheduled_tokens[i] for i in req_ids]
+    num_scheduled_tokens = np.array(tokens, dtype=np.int32)
+    max_num_scheduled_tokens = max(tokens)
+
+    # Get request indices.
+    # E.g., [2, 5, 3] -> [0, 0, 1, 1, 1, 1, 1, 2, 2, 2]
+    req_indices = np.repeat(runner.arange_np[:num_reqs],
+                            num_scheduled_tokens) + req_offset
+
+    # cu_num_tokens: [2, 5, 3] -> [2, 7, 10]
+    # arange: [0, 1, 0, 1, 2, 3, 4, 0, 1, 2]
+    cu_num_tokens, arange = runner._get_cumsum_and_arange(
+        num_scheduled_tokens)
+
+    # Get positions.
+    positions_np = runner.positions_np[:total_num_scheduled_tokens]
+    np.add(runner.input_batch.num_computed_tokens_cpu[req_indices],
+            arange,
+            out=positions_np)
+
+    # Calculate the slot mapping for each KV cache group.
+    for kv_cache_group_id, kv_cache_group_spec in enumerate(
+            runner.kv_cache_config.kv_cache_groups):
+        block_size = kv_cache_group_spec.kv_cache_spec.block_size
+        block_table: BlockTable = runner.input_batch.block_table[
+            kv_cache_group_id]
+        # E.g., [0, 1, 0, 1, 2, 3, 4, 0, 1, 2]
+        # -> [0, 0, K, K, K + 1, K + 1, K + 2, 2 * K, 2 * K, 2 * K + 1]
+        # where K is the max_num_blocks_per_req and the block size is 2.
+        # NOTE(woosuk): We can't simply use `token_indices // block_size`
+        # here because M (max_model_len) is not necessarily divisible by
+        # block_size.
+        block_table_indices = (
+            req_indices * block_table.max_num_blocks_per_req +
+            positions_np // block_size)
+        block_table_cpu = block_table.get_cpu_tensor()
+        block_numbers = block_table_cpu.flatten(
+        )[block_table_indices].numpy()
+        block_offsets = positions_np % block_size
+        np.add(
+            block_numbers * block_size,
+            block_offsets,
+            out=block_table.slot_mapping_np[:total_num_scheduled_tokens])
+
+    # Prepare the attention metadata.
+    runner.query_start_loc_np[0] = 0
+    runner.query_start_loc_np[1:num_reqs + 1] = cu_num_tokens
+
+    runner.seq_lens_np[:num_reqs] = (
+        runner.input_batch.num_computed_tokens_cpu[req_offset : req_offset + num_reqs] +
+        num_scheduled_tokens)
+    runner.query_start_loc[query_start_offset: query_start_offset + num_reqs + 1].copy_(
+        runner.query_start_loc_cpu[:num_reqs + 1], non_blocking=True)
+    # Note: pad query_start_loc to be non-decreasing, as kernels
+    # like FlashAttention requires that
+    if req_offset > 0: #right
+        runner.query_start_loc[query_start_offset + num_reqs + 1:].fill_(
+            runner.query_start_loc_cpu[num_reqs].item())
+    runner.seq_lens[seq_len_offset :seq_len_offset + num_reqs].copy_(runner.seq_lens_cpu[:num_reqs],
+                                    non_blocking=True)
+
+    # Fill unused with -1. Needed for reshape_and_cache
+    if req_offset > 0: #right
+        runner.seq_lens[seq_len_offset + num_reqs:].fill_(0)
+
+    query_start_loc = runner.query_start_loc[query_start_offset: query_start_offset + num_reqs + 1]
+    seq_lens = runner.seq_lens[seq_len_offset : seq_len_offset + num_reqs]
+
+    common_attn_metadata = CommonAttentionMetadata(
+        query_start_loc=query_start_loc, 
+        seq_lens=seq_lens,
+        num_reqs=num_reqs,
+        num_actual_tokens=total_num_scheduled_tokens,
+        max_query_len=max_num_scheduled_tokens)
+
+    attn_metadata: dict[str, Any] = {}
+    # Prepare the attention metadata for each KV cache group and make layers
+    # in the same group share the same metadata.
+    for kv_cache_group_id, kv_cache_group_spec in enumerate(
+            runner.kv_cache_config.kv_cache_groups):
+
+        # Prepare for cascade attention if enabled & beneficial.
+        common_prefix_len = 0
+        if runner.cascade_attn_enabled:
+            common_prefix_len = runner._compute_cascade_attn_prefix_len(
+                num_scheduled_tokens,
+                scheduler_output.
+                num_common_prefix_blocks[kv_cache_group_id],
+                kv_cache_group_spec.kv_cache_spec,
+                runner.attn_metadata_builders[kv_cache_group_id],
+            )
+        if req_offset > 0:
+            origin_block_table = runner.attn_metadata_builders[kv_cache_group_id].block_table.block_table
+            runner.attn_metadata_builders[kv_cache_group_id].block_table.block_table = origin_block_table[req_offset:, :]
+            origin_slot_mapping = runner.attn_metadata_builders[kv_cache_group_id].block_table.slot_mapping
+            runner.attn_metadata_builders[kv_cache_group_id].block_table.slot_mapping = \
+                origin_slot_mapping[input_split.scheduler_output_left.total_num_scheduled_tokens:]
+
+        attn_metadata_i = (
+            runner.attn_metadata_builders[kv_cache_group_id].build(
+                common_prefix_len=common_prefix_len,
+                common_attn_metadata=common_attn_metadata)) # maybe FlashAttentionMetadata
+        if req_offset > 0:
+            runner.attn_metadata_builders[kv_cache_group_id].block_table.block_table = origin_block_table
+            runner.attn_metadata_builders[kv_cache_group_id].block_table.slot_mapping = origin_slot_mapping
+        for layer_name in kv_cache_group_spec.layer_names:
+            attn_metadata[layer_name] = attn_metadata_i
+
+    return attn_metadata
+
+def pad_num_input_tokens(self, scheduler_output):
+    num_scheduled_tokens = scheduler_output.total_num_scheduled_tokens
+    if (self.use_cuda_graph
+            and num_scheduled_tokens <= self.cudagraph_batch_sizes[-1]):
+        # Use piecewise CUDA graphs.
+        # Add padding to the batch size.
+        num_input_tokens = self.vllm_config.pad_for_cudagraph(
+            num_scheduled_tokens)
+    else:
+        # Eager mode.
+        # Pad tokens to multiple of tensor_parallel_size when
+        # enabled collective fusion for SP
+        tp_size = self.vllm_config.parallel_config.tensor_parallel_size
+        if self.vllm_config.compilation_config.pass_config. \
+            enable_sequence_parallelism and tp_size > 1:
+            from vllm.utils import round_up
+            num_input_tokens = round_up(num_scheduled_tokens, tp_size)
+        else:
+            num_input_tokens = num_scheduled_tokens
+
+    # Padding for DP
+    num_pad, num_tokens_across_dp = self.get_dp_padding(num_input_tokens)
+    num_input_tokens += num_pad
+    return num_input_tokens, num_tokens_across_dp
+
+def tbo_split_and_execute_model(
+    runner,
+    attn_metadata,
+    num_input_tokens,
+    num_tokens_across_dp,
+    input_ids,
+    positions,
+    inputs_embeds,
+    scheduler_output: "SchedulerOutput",
+    intermediate_tensors: Optional[IntermediateTensors] = None,
+) -> Union[ModelRunnerOutput, IntermediateTensors]:
+    use_tbo = False
+    if len(scheduler_output.num_scheduled_tokens) > 1:
+        split_scheduler_output(runner, scheduler_output)
+        if input_split.scheduler_output_left.total_num_scheduled_tokens >= envs.VLLM_TBO_MIN_TOKENS and \
+            input_split.scheduler_output_right.total_num_scheduled_tokens >= envs.VLLM_TBO_MIN_TOKENS:
+            use_tbo = True
+
+    if use_tbo:
+        num_input_tokens_left = input_split.scheduler_output_left.total_num_scheduled_tokens
+        num_input_tokens_right = num_input_tokens - num_input_tokens_left
+        
+        attn_metadata_left = prepare_tbo_atten_metadata(runner, input_split.scheduler_output_left, input_split.req_ids_left, 0)
+        attn_metadata_right = prepare_tbo_atten_metadata(runner, input_split.scheduler_output_right, input_split.req_ids_right, input_split.req_num_left)
+
+        model_output = tbo_model_executable_v1(
+            runner,
+            attn_metadata_left, 
+            attn_metadata_right, 
+            num_input_tokens_left,
+            num_input_tokens_right,
+            num_tokens_across_dp,
+            input_ids,
+            positions,
+            intermediate_tensors,
+            inputs_embeds) 
+        finished_sending, finished_recving = None, None
+    else:
+        # Run the decoder.
+        # Use persistent buffers for CUDA graphs.
+        with set_forward_context(attn_metadata,
+                                runner.vllm_config,
+                                num_tokens=num_input_tokens,
+                                num_tokens_across_dp=num_tokens_across_dp):
+            runner.maybe_setup_kv_connector(scheduler_output)
+
+            model_output = runner.model(
+                input_ids=input_ids,
+                positions=positions,
+                intermediate_tensors=intermediate_tensors,
+                inputs_embeds=inputs_embeds,
+            )
+
+            runner.maybe_wait_for_kv_save()
+            finished_sending, finished_recving = (
+                runner.get_finished_kv_transfers(scheduler_output))
+    return model_output, finished_sending, finished_recving
\ No newline at end of file

vllm/v1/worker/gpu_model_runner.py

@@ -68,6 +68,7 @@ from vllm.v1.worker.block_table import BlockTable
 from vllm.v1.worker.gpu_input_batch import CachedRequestState, InputBatch
 from vllm.v1.worker.lora_model_runner_mixin import LoRAModelRunnerMixin
 from vllm.platforms import current_platform
+from vllm.two_batch_overlap.v1.model_input_split_v1 import tbo_split_and_execute_model
 
 from ..sample.logits_processor import LogitsProcessorManager
 from .utils import (gather_mm_placeholders, initialize_kv_cache_for_kv_sharing,
@@ -1361,6 +1362,12 @@ class GPUModelRunner(LoRAModelRunnerMixin):
         # compiled with full CUDA graphs, we have to skip them entirely.
         skip_cuda_graphs = self.full_cuda_graph and not attention_cuda_graphs
 
+        if envs.VLLM_ENABLE_TBO and not self.use_cuda_graph:
+            model_output, finished_sending, finished_recving = \
+                 tbo_split_and_execute_model(self, attn_metadata, num_input_tokens,
+                                             num_tokens_across_dp, input_ids, positions,
+                                             inputs_embeds, scheduler_output, intermediate_tensors)
+        else:
             # Run the model.
             # Use persistent buffers for CUDA graphs.
             with set_forward_context(

vllm/v1/worker/gpu_worker.py

@@ -22,7 +22,6 @@ from vllm.lora.request import LoRARequest
 from vllm.model_executor import set_random_seed
 from vllm.platforms import current_platform
 from vllm.sequence import IntermediateTensors
-from vllm.two_batch_overlap.v1.gpu_model_runner import TBO_GPUModelRunner
 from vllm.utils import GiB_bytes, MemorySnapshot, memory_profiling
 from vllm.v1.kv_cache_interface import KVCacheConfig, KVCacheSpec
 from vllm.v1.outputs import ModelRunnerOutput
@@ -163,10 +162,6 @@ class Worker(WorkerBase):
         set_random_seed(self.model_config.seed)
 
         # Construct the model runner
-        if envs.VLLM_ENABLE_TBO:
-            self.model_runner: TBO_GPUModelRunner = TBO_GPUModelRunner(
-                self.vllm_config, self.device)
-        else:
         self.model_runner: GPUModelRunner = GPUModelRunner(
             self.vllm_config, self.device)