Skip to content

[https://nvbugs/5687820][fix] Remove self.abort() in DetokenizedGenerationResult #9449

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
syuoni merged 1 commit into from
Nov 27, 2025
+11 −3
Merged

[https://nvbugs/5687820][fix] Remove self.abort() in DetokenizedGenerationResult #9449

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions examples/disaggregated/slurm/benchmark/config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,6 +76,7 @@ worker_config:
- 2048
- 256
print_iter_log: true
trust_remote_code: true
kv_cache_config:
enable_block_reuse: false
free_gpu_memory_fraction: 0.8
Expand All @@ -102,6 +103,7 @@ worker_config:
enable_attention_dp: true
pipeline_parallel_size: 1
print_iter_log: true
trust_remote_code: true
cuda_graph_config: null
disable_overlap_scheduler: true
kv_cache_config:
Expand Down
1 change: 1 addition & 0 deletions examples/disaggregated/slurm/benchmark/run_benchmark.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@ for concurrency in ${concurrency_list}; do
--dataset-path ${dataset_file} \
--num-prompts ${num_prompts} \
--max-concurrency ${concurrency} \
--trust-remote-code \
--ignore-eos \
--no-test-input \
--save-result \
Expand Down
1 change: 1 addition & 0 deletions examples/disaggregated/slurm/benchmark/run_benchmark_nv_sa.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,6 +75,7 @@ for concurrency in ${concurrency_list}; do
--dataset-name random \
--num-prompts "${num_prompts}" \
--max-concurrency "${concurrency}" \
--trust-remote-code \
--ignore-eos \
--random-input-len "${input_seq_len}" \
--random-output-len "${output_seq_len}" \
Expand Down
9 changes: 7 additions & 2 deletions examples/disaggregated/slurm/benchmark/submit.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,8 +105,13 @@ def submit_job(config):
log_base = os.path.join(env_config['work_dir'], f"{isl}-{osl}")

# Get eplb num_slots for gen worker
eplb_num_slots = (config['worker_config']['gen'].get('moe_config', {}).get(
'load_balancer', {}).get('num_slots', 0))
load_balancer_config = config['worker_config']['gen'].get(
'moe_config', {}).get('load_balancer', {})
if isinstance(load_balancer_config, str):
with open(load_balancer_config, 'r') as f:
load_balancer_config = yaml.safe_load(f)
eplb_num_slots = load_balancer_config.get('num_slots', 0)

# Determine directory suffix based on attention_dp
if gen_enable_attention_dp:
dir_suffix = f"ctx{ctx_num}_gen{gen_num}_dep{gen_tp_size}_batch{gen_batch_size}_eplb{eplb_num_slots}_mtp{mtp_size}"
Expand Down
1 change: 0 additions & 1 deletion tensorrt_llm/executor/result.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -761,7 +761,6 @@ def _handle_response(self, response: "GenerationExecutor.Response"):

beam_output.finish_reason = 'stop'
beam_output.stop_reason = stop_reason
self.abort()
self._done = True
break

Expand Down