Go to file

claude 3d816cd18d fix: add ceil_div + DEVICE_INLINE to device_utils.cuh

ceil_div<T> was in xllm utils.h (removed for glog).
DEVICE_INLINE macro also moved to shared header.

2026-08-14 11:11:15 +00:00

cccl_upstream

…

docs

…

ex_engine

fix: add ceil_div + DEVICE_INLINE to device_utils.cuh

2026-08-14 11:11:15 +00:00

ixformer_sdk

…

muh

…

optimizations

…

qwen3_6_scripts

…

upstream_ref

…

vllm

…

vllm_adapter

…

vllm_overrides

…

__init__.py

…

.dockerignore

…

.gitattributes

…

.gitignore

…

attention.py

…

baseline.muh

…

BI_V100_BENCHMARK_RUNBOOK.md

…

cat_ixformer_vllm.py

…

CCCL_ASSET_MAP.md

…

CCCL_INTEGRATION_STATUS.md

…

CCCL_MUH_GAP_ANALYSIS.md

…

CCCL_MUH_PARITY_AUDIT.md

…

CCCL_PATTERN_MAP.md

…

cccl_sm100_benchmark_values.json

…

CCCL_TUNING_GAP_REPORT.md

…

chat_dataset_v0.json

…

CODEPATH_MAP.md

…

COMP168_DIAGNOSIS.md

…

COMPETITIVE_ANALYSIS_AND_FIX_PLAN.md

…

computility-run.fix.yaml

…

computility-run.ref.yaml

…

computility-run.yaml

…

computility-run.yaml.bak

…

debug_gdn_nan.py

…

debug_topk.py

…

debug_warpsize.py

…

deltanet_chunk_optimize.py

…

DEVELOPMENT_STATUS.md

…

diagnose_build.sh

…

DLOPEN_DEV_PLAN.md

…

DLOPEN_DISPATCH_CHAIN.md

…

Dockerfile

…

Dockerfile.broken_head

…

Dockerfile.broken_head2

…

Dockerfile.fix

…

Dockerfile.ref

…

dockerrizhi.txt

…

engine_cccl_patterns.py

…

ENGINE_CODEPATH_TIMELINE.md

…

ENGINEX_INJECTION_MAP.md

…

enginex-vllm-bi100-qwen36-main.zip

…

GROUND_TRUTH_STATUS_v2.md

…

GROUND_TRUTH_STATUS.md

…

HARDWARE_PROBE_20260808.md

…

launch_service

…

muh_cc_dispatch.py

…

muh_dispatch.py

…

muh_kernel_map.py

…

MUH_PROJECT_CHECKPOINT.md

…

MUH_TUNING_GAP_ANALYSIS.md

…

paged_attention_v2_pytorch.py

…

paged_attention_v2_triton.py

…

paged_attn.py

…

PIPELINE_GROUND_TRUTH.md

…

PIPELINE_REALITY_CHECK.md

…

PIPELINE_STATUS.md

…

PRD.md

…

prefix_prefill.py

…

probe_all_so.py

…

probe_all_symbols.sh

…

probe_base_moe.py

…

probe_bi100.py

…

probe_ixformer_symbols.py

…

probe_kv_layout2.py

…

probe_kv_layout.py

…

probe_moe_detail.py

…

probe_paged_attn.py

…

probe_symbol.sh

…

PROJECT_SUMMARY.md

…

README.md

…

SUB509_DEEP_DIAGNOSIS.md

…

SUB509_DIAGNOSIS.md

…

subrizhi.txt

…

SYSTEM_DESIGN.md

…

test_dlopen_chain.py

…

test_triton.py

…

TUNING_SURFACE_TRUTH.md

…

verify_build.sh

…

verify_deployment.sh

…

verify_docker_build.sh

…

verify_flash_attn.py

…

verify_forward.py

…

verify_gdn_cpp.py

…

verify_ixformer_attn.py

…

verify_moe_e2e.py

…

verify_moe_index_combine.py

…

verify_on_device.py

…

verify_paged_attn.py

…

verify_single_card.py

…

verify_single_gpu.py

…

verify_so_loading.py

…

verify_submission.sh

…

verify_topk_softmax.py

…

README.md

project_6

Languages

C++ 41.8%

Cuda 31.6%

Python 22.2%

C 2.1%

CMake 1.1%

Other 1.1%