project_6

dylanyunlong/project_6

Fork 0

Commit Graph

Author	SHA1	Message	Date
Claude	eb57eb7d1c	clean: remove 444 .pyc files + libcccl_allocator.so from git tracking These cause docker build failures on competition platform. .gitignore and .dockerignore already exclude them.	2026-08-14 02:20:26 +00:00
Dylan	d15dcea7c6	[ENGINE] port SDPA fallback for head_dim>128 to base xformers backend Source: qwen3_6_scripts/xformers.py (competition-specific) CCCL ref: agent_reduce.cuh ConsumeFullTile (GQA broadcast) block_load_to_shared.cuh (loop invariant hoisting) agent_sub_warp_merge_sort.cuh (buffer reuse) CRITICAL: Qwen3.6 uses head_dim=256. ixformer flash attention only supports head_dim<=128. Without this fallback, base xformers.py would try ixformer flash on head_dim=256 -> crash or wrong results. SDPA fallback features (CCCL-driven): 1. Q-tiling with _Q_CHUNK=256: O(chunk*seq) memory, not O(seq^2) 2. GQA broadcast matmul: K/V as [kv_h,1,seq,d], broadcast over gqa groups -> 6x memory savings vs repeat_interleave for Qwen3.6 3. Pre-allocated loop invariants (k_pos, qc_q_pos_base) 4. Float32 softmax to prevent fp16 overflow This directly impacts all 50+ functional test cases that use prefill.	2026-08-07 01:24:04 +00:00
dylanyunlon	ef6abf3dc7	[DEPLOY] Complete submission: baseline + all optimizations Adds ALL files needed for Dockerfile build: - qwen3_6_scripts/ (baseline patches + our optimizations) - vllm/ (full vllm package) - paged_attention_v2_pytorch.py (V2 with single-bmm optimization) - Dockerfile + computility-run.yaml Our optimizations vs baseline: 1. paged_attn.py: pre-gathered context KV (eliminates 194 gather calls), Triton try/fallback, V2 heuristic, threshold 32K→64K 2. paged_attention_v2_pytorch.py: fills NotImplementedError, single-bmm Phase 1 (195 launches → 3) 3. patch_enable_triton.py: HAS_TRITON=True with safety fallback 4. patch_triton_tuning.py: BLOCK=64, NUM_WARPS=4 for BI-V100 5. computility-run.yaml: gpu-memory-utilization 0.9→0.95, max-num-batched-tokens 8192→16384 This repo can now be submitted to dev.modelhub.org.cn as-is.	2026-07-30 16:06:20 +00:00

Author

SHA1

Message

Date

Claude

eb57eb7d1c

clean: remove 444 .pyc files + libcccl_allocator.so from git tracking

These cause docker build failures on competition platform.
.gitignore and .dockerignore already exclude them.

2026-08-14 02:20:26 +00:00

Dylan

d15dcea7c6

[ENGINE] port SDPA fallback for head_dim>128 to base xformers backend

Source: qwen3_6_scripts/xformers.py (competition-specific)
CCCL ref: agent_reduce.cuh ConsumeFullTile (GQA broadcast)
          block_load_to_shared.cuh (loop invariant hoisting)
          agent_sub_warp_merge_sort.cuh (buffer reuse)

CRITICAL: Qwen3.6 uses head_dim=256. ixformer flash attention only
supports head_dim<=128. Without this fallback, base xformers.py would
try ixformer flash on head_dim=256 -> crash or wrong results.

SDPA fallback features (CCCL-driven):
1. Q-tiling with _Q_CHUNK=256: O(chunk*seq) memory, not O(seq^2)
2. GQA broadcast matmul: K/V as [kv_h,1,seq,d], broadcast over gqa
   groups -> 6x memory savings vs repeat_interleave for Qwen3.6
3. Pre-allocated loop invariants (k_pos, qc_q_pos_base)
4. Float32 softmax to prevent fp16 overflow

This directly impacts all 50+ functional test cases that use prefill.

2026-08-07 01:24:04 +00:00

dylanyunlon

ef6abf3dc7

[DEPLOY] Complete submission: baseline + all optimizations

Adds ALL files needed for Dockerfile build:
  - qwen3_6_scripts/ (baseline patches + our optimizations)
  - vllm/ (full vllm package)
  - paged_attention_v2_pytorch.py (V2 with single-bmm optimization)
  - Dockerfile + computility-run.yaml

Our optimizations vs baseline:
  1. paged_attn.py: pre-gathered context KV (eliminates 194 gather calls),
     Triton try/fallback, V2 heuristic, threshold 32K→64K
  2. paged_attention_v2_pytorch.py: fills NotImplementedError,
     single-bmm Phase 1 (195 launches → 3)
  3. patch_enable_triton.py: HAS_TRITON=True with safety fallback
  4. patch_triton_tuning.py: BLOCK=64, NUM_WARPS=4 for BI-V100
  5. computility-run.yaml: gpu-memory-utilization 0.9→0.95,
     max-num-batched-tokens 8192→16384

This repo can now be submitted to dev.modelhub.org.cn as-is.

2026-07-30 16:06:20 +00:00

3 Commits