dylanyunlong/project_6
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Commit Graph

  • f8153d492a [docs] agent_scan.cuh architecture — scan SMEM is real, reduce SMEM is not project_6 2026-08-05 03:33:38 +00:00
  • fa8a02c642 Merge branch 'main' of https://github.com/dylanyunlon/project_6 muh-bot 2026-08-05 03:32:38 +00:00
  • 41ecb8c9e9 [analysis] 关键架构发现: gen_patch 注入目标不存在 + paged_attn V1 硬编码 muh-bot 2026-08-05 03:32:31 +00:00
  • 44e4f6f947 [v2] PARTITION_SIZE 512→1024 + fix import path project_6 2026-08-05 03:32:23 +00:00
  • a0cddf2ddc Merge branch 'main' of https://github.com/dylanyunlon/project_6 muh-bot 2026-08-05 03:26:40 +00:00
  • 8d26e23e8e [muh] fix scale_reg: 补上 CCCL scale_reg_bound 的 threads SMEM cap muh-bot 2026-08-05 03:26:36 +00:00
  • 33e1a21a66 [v2] Wire paged_attention_v2_pytorch into vllm — enable V2 for long sequences project_6 2026-08-05 03:26:18 +00:00
  • faedfab7e9 Merge branch 'main' of https://github.com/dylanyunlon/project_6 muh-bot 2026-08-05 03:22:21 +00:00
  • 23e34fde33 [muh] tuning_radix_sort: 148→211行, 基于 CCCL 2381 行源码完整重建 10-子策略架构 muh-bot 2026-08-05 03:22:17 +00:00
  • 3cc97c1d4e [docs] CCCL scan architecture — lookback vs lookahead, tile_state allocation, grid sizing project_6 2026-08-05 03:21:47 +00:00
  • b50bd2dfd5 [gen_patch] fix critical struct selection: dispatch by kernel data type muh-bot 2026-08-05 03:20:39 +00:00
  • 55b704c0e0 [docs] CCCL reduce architecture deep dive — SMEM model correction + V1/V2 dispatch finding project_6 2026-08-05 03:20:51 +00:00
  • 2795d2b7f2 [muh] tuning_reduce: 修正 SMEM 模型错误 (基于 agent_reduce.cuh 源码分析) muh-bot 2026-08-05 03:20:38 +00:00
  • 6bf73bdacb [moe] BLOCK_SIZE_M heuristic refined for BI-V100 decode workload project_6 2026-08-05 03:17:53 +00:00
  • 5379a573ac [yaml+prefill] num-scheduler-steps 8→16 from CCCL delay analysis project_6 2026-08-05 03:16:01 +00:00
  • db0df78580 [gen_patch] fix 3 critical bugs: reduce struct selection, topk bits_per_pass injection, transform/batch_memcpy extraction project_6 2026-08-05 03:15:12 +00:00
  • 2c43eb524f [flash_attn] CCCL-derived autotune configs: num_stages=2 + small-tile project_6 2026-08-05 03:09:45 +00:00
  • 8a87e378f8 [prefill] asymmetric BLOCK_M/BLOCK_N from CCCL AgentReduce insight project_6 2026-08-05 03:07:14 +00:00
  • 8a38c04b4c [vllm] 3 个运行时 bug 修复: SMEM 32KB→48KB, NUM_WARPS 8→4, v2 归一化 dylanyunlon 2026-08-04 12:26:24 +00:00
  • 11032fe95e [muh] delay v2 完成: 全部 8 个 lookback 算法改为 no_delay muh-bot 2026-08-04 07:18:48 +00:00
  • a898faa34e [muh] delay v2: reduce_by_key + scan_by_key 全部改为 no_delay muh-bot 2026-08-04 07:17:34 +00:00
  • 31d39e6032 [muh] 首批 BI-V100 实测数据写入 3 个 tuning headers: scan/topk/transform dylanyunlon 2026-08-04 07:13:01 +00:00
  • 475574fd3d [muh] bench_triton_real + bi100_configs: REAL tunable surface benchmark project_6 2026-08-04 06:19:07 +00:00
  • 5747e7c290 [docs] BI-V100 benchmark runbook: Phase 0 硬件探测 + Phase 1 quick sweep + Phase 3 端到端验证 Claude 2026-08-04 01:14:30 +00:00
  • 12ad7a3190 [muh] 7 headers 完整移植 CCCL tuning tables: segmented_sort 7%→29%, merge_sort 22%→43%, merge 30%→49%, adjacent_difference 38%→65%, batch_memcpy 37%→41%, find 35%→43%, find_bound 31%→44% dylanyunlon 2026-08-03 21:36:46 +00:00
  • 2badbfa1b9 merge: resolve conflicts, keep full CCCL port versions muh-bot 2026-08-03 21:35:31 +00:00
  • c7ff12c28d [muh] scan_by_key 14%→13%, rle_non_trivial_runs 7%→10%, rle_encode 9%→10%: 从 CCCL 3325 行源码完整移植 muh-bot 2026-08-03 21:35:16 +00:00
  • 6a56649d9a [muh] radix_sort 6%→19%, rle_encode 8%→21%, rle_non_trivial_runs 6%→18%: 完整移植 CCCL SM90/SM100 tuning tables + BI-V100 SMEM 48KB 约束 dylanyunlon 2026-08-03 13:14:54 +00:00
  • 1b74226910 [muh] reduce_by_key 12%→16%: 从 CCCL 1735 行源码完整移植 66 条 SM80/SM90/SM100 tuning entries muh-bot 2026-08-03 13:11:31 +00:00
  • 5923223cba [muh] three_way_partition 7%→13%, histogram 13%→21%: 继续从 CCCL 源码移植 SM80/SM90/SM100 tuning tables muh-bot 2026-08-03 13:02:58 +00:00
  • c350c1c7e5 [muh] select_if 5%→17%, unique_by_key 3%→11%: 从 CCCL 源码完整移植 SM80/SM90/SM100 三代 tuning table muh-bot 2026-08-03 13:01:11 +00:00
  • 24ef6a91b5 [CCCL] 瘦身 + 补全: 移除 cudax/python/libcudacxx-tests 冗余文件, 新增 c2h 测试助手 + cmake 构建系统 + 8 个 CUDA thrust examples muh-bot 2026-08-03 12:39:26 +00:00
  • a2a5dd8f00 feat: asymmetric BLOCK_M/BLOCK_N search + re-add BI-V100 autotune configs Claude 2026-08-03 11:18:18 +00:00
  • fe64650681 add muh/run_on_bi100.sh — single-paste diagnostic + benchmark for Phanthy Cloud dylanyunlon 2026-08-03 10:51:20 +00:00
  • 327497dc17 feat(muh): add bench_triton_prefill.py — real compile-time param injection benchmark dylanyunlon 2026-08-03 10:46:19 +00:00
  • 8c1955dc92 fix: revert invalid patches, add honest tuning surface assessment dylanyunlon 2026-08-03 10:34:28 +00:00
  • dc9ac0a757 feat(muh): apply CCCL-derived BI-V100 tuning to 5 vllm Python files dylanyunlon 2026-08-03 10:27:10 +00:00
  • 094c710efa feat(muh): add bench_bi100.py — CCCL BruteForceSeeker for BI-V100 dylanyunlon 2026-08-03 10:24:37 +00:00
  • 9f93d695a9 feat: deploy CCCL-tuned prefix_prefill + muh_dispatch + fix SM=16 count Claude 2026-08-03 08:30:16 +00:00
  • 8e9c22f6c1 feat: CCCL-derived 3-tier decode dispatch + SM=16 prefill tuning + multi-step scheduling Claude 2026-08-03 08:28:38 +00:00
  • 16981f221e feat(muh): reduce_by_key 55→203 lines — full key_size×accum_size dispatch Claude 2026-08-03 07:20:23 +00:00
  • bdaec8da4b feat(muh): select_if SM=16 tile maximization Claude 2026-08-03 07:19:13 +00:00
  • 95d872e8f5 feat(muh): scan_by_key 53→284 lines — full key_size×val_size type dispatch Claude 2026-08-03 07:18:23 +00:00
  • 88db0ed89c feat(muh): SM=16 tuning overhaul — reduce/scan/transform Claude 2026-08-03 07:16:35 +00:00
  • 0ba4cdb025 fix: dial back max-num-seqs 256→8, revert batched-tokens and mem-util Claude 2026-08-03 06:51:33 +00:00
  • cdc01bbc6a fix: critical config + tuning corrections from CCCL source analysis Claude 2026-08-03 06:45:54 +00:00
  • 071fa361a3 docs: CCCL → Triton methodology transfer — parameter search translation table Claude 2026-08-03 04:37:23 +00:00
  • e48a46a30d docs: EngineX vllm injection map — Python not C++, Triton not CUDA, 32KB SMEM claim Claude 2026-08-03 03:59:49 +00:00
  • 9bba7f4c79 docs: CCCL integration status — 462 assets inventory, 8/8 decode hot path coverage, no-clone-more verdict Claude 2026-08-03 03:50:24 +00:00
  • 392e644611 feat: add qwen3_5.py vllm adapter for Qwen3.6-35B-A3B Claude 2026-08-01 13:54:25 +00:00
  • 6beb497447 fix(hardware): SM count 50→16 confirmed on Phanthy Cloud BI-V100 root 2026-08-01 13:39:28 +00:00
  • fee8f1b9e4 docs: document Qwen3.6-35B-A3B bootstrap failure and architecture analysis Claude 2026-08-01 13:16:01 +00:00
  • 9b0d1c283c docs: add competition server profile (4×BI-V100, Qwen3.6-35B-A3B) Claude 2026-08-01 13:12:59 +00:00
  • 79730ea907 test: add SMEM safety validator for all 26 tuning algorithms dylanyunlon 2026-08-01 12:37:14 +08:00
  • 0154a3b297 fix(tuning_batched_topk): force bits=8, fix SMEM overflow dylanyunlon 2026-08-01 12:36:57 +08:00
  • 2c5e77f370 feat(gen_patch): add TUNING_REGISTRY for all 26 algorithms dylanyunlon 2026-08-01 12:33:14 +08:00
  • 84c18150e6 fix(tuning_select_if): restore 3 collapsed dispatch dimensions dylanyunlon 2026-08-01 02:26:48 +08:00
  • 9287700964 fix(tuning_radix_sort): remove invented portioned_smem_per_warp field dylanyunlon 2026-08-01 02:26:46 +08:00
  • 2bc3263793 [muh] add tuning_radix_sort.cuh: BI-V100 tuning for radix_sort dylanyunlon 2026-08-01 02:11:05 +08:00
  • 437fc3ea20 [muh] add tuning_unique_by_key.cuh: BI-V100 tuning for unique_by_key dylanyunlon 2026-08-01 02:11:03 +08:00
  • 58de86d817 [muh] add tuning_select_if.cuh: BI-V100 tuning for select_if dylanyunlon 2026-08-01 02:11:01 +08:00
  • 91f9a3a0e5 [muh] add tuning_scan_by_key.cuh: BI-V100 tuning for scan_by_key dylanyunlon 2026-08-01 02:10:59 +08:00
  • 0ec355cf74 [muh] add tuning_reduce_by_key.cuh: BI-V100 tuning for reduce_by_key dylanyunlon 2026-08-01 02:10:57 +08:00
  • 915c4aff56 [muh] add tuning_segmented_sort.cuh: BI-V100 tuning for segmented_sort dylanyunlon 2026-08-01 02:10:55 +08:00
  • 561a82c849 [muh] add tuning_three_way_partition.cuh: BI-V100 tuning for three_way_partition dylanyunlon 2026-08-01 02:10:54 +08:00
  • aeb270578f [muh] add tuning_rle_non_trivial_runs.cuh: BI-V100 tuning for rle_non_trivial_runs dylanyunlon 2026-08-01 02:10:52 +08:00
  • f3ae28bbb5 [muh] add tuning_rle_encode.cuh: BI-V100 tuning for rle_encode dylanyunlon 2026-08-01 02:10:50 +08:00
  • f838c22bad [muh] add tuning_histogram.cuh: BI-V100 tuning for histogram dylanyunlon 2026-08-01 02:10:48 +08:00
  • 3a8030224b [muh] add tuning_segmented_radix_sort.cuh: BI-V100 tuning header for segmented_radix_sort dylanyunlon 2026-08-01 02:08:44 +08:00
  • c3e9e5b6f5 [muh] add tuning_batched_topk.cuh: BI-V100 tuning header for batched_topk dylanyunlon 2026-08-01 02:08:42 +08:00
  • 0344a3fbb8 [muh] add tuning_transform_tile.cuh: BI-V100 tuning header for transform_tile dylanyunlon 2026-08-01 02:08:40 +08:00
  • 3fee54f4d7 [muh] add tuning_merge_sort.cuh: BI-V100 tuning header for merge_sort dylanyunlon 2026-08-01 02:08:38 +08:00
  • af8fc0caeb [muh] add tuning_merge.cuh: BI-V100 tuning header for merge dylanyunlon 2026-08-01 02:08:37 +08:00
  • c0bfc8c93d [muh] add tuning_segmented_scan.cuh: BI-V100 tuning header for segmented_scan dylanyunlon 2026-08-01 02:08:35 +08:00
  • 105dd96b52 [muh] add tuning_segmented_reduce.cuh: BI-V100 tuning header for segmented_reduce dylanyunlon 2026-08-01 02:08:33 +08:00
  • 81e4a907c0 [muh] add tuning_find_bound_sorted_values.cuh: BI-V100 tuning header for find_bound_sorted_values dylanyunlon 2026-08-01 02:08:31 +08:00
  • 25f7a636a9 [muh] add tuning_find.cuh: BI-V100 tuning header for find dylanyunlon 2026-08-01 02:08:29 +08:00
  • eaca73a390 [muh] add tuning_adjacent_difference.cuh: BI-V100 tuning header for adjacent_difference dylanyunlon 2026-08-01 02:08:28 +08:00
  • 52c5ca7ce5 refactor(muh_dispatch): read-once from C++ headers, not write-twice dylanyunlon 2026-08-01 01:30:42 +08:00
  • 482aabdea3 fix(muh_kernel_map): add threads >= 32 floor in Python scale_mem_bound dylanyunlon 2026-08-01 01:29:57 +08:00
  • 142568072a fix(common.cuh): add threads >= 32 floor in scale_mem_bound dylanyunlon 2026-08-01 01:29:47 +08:00
  • 03f6a59ebf fix(muh_dispatch): add verify_against_headers() to close the loop dylanyunlon 2026-08-01 00:32:16 +08:00
  • 3a2b67c166 fix(tuning_reduce): auto [t,i] → auto [i,t] matching CCCL scaling_result dylanyunlon 2026-08-01 00:31:41 +08:00
  • ec1c85cd9a fix(common.cuh): scale_mem_bound — 3 bugs vs CCCL original dylanyunlon 2026-08-01 00:31:22 +08:00
  • 3ebc37d80d [muh] fix scale_mem_bound: 3 bugs vs CCCL util_arch.cuh dylanyunlon 2026-08-01 00:00:02 +08:00
  • 173c6afe09 [muh] kernel_map: full vllm→CCCL mapping with SMEM overflow detection Claude 2026-07-31 11:13:33 +00:00
  • e69c46d0b7 [muh] add muh_dispatch.py — CCCL-style type-dispatched kernel config for BI-V100 dylanyunlon 2026-07-31 19:11:46 +08:00
  • d14b0c19e4 [docs] add CCCL SM100 vs muh BI-V100 specialization parity analysis dylanyunlon 2026-07-31 18:35:50 +08:00
  • 35ef79c5f8 [muh] scan: add bi100_lookback_1B_o8 — close SM100 parity gap (7/7 lookback branches) dylanyunlon 2026-07-31 18:35:01 +08:00
  • c5a0d61851 sync: align with enginex-vllm-bi100-qwen36 baseline (1902c81f) Claude 2026-07-31 09:43:58 +00:00
  • de7ee4383e [VERIFIED] Hardware-tested native kernel integration Claude 2026-07-31 06:43:25 +00:00
  • 78a0ebd516 [CRITICAL] Fix V2 cache layout: V1=5D K, V2=4D K with transposed layout Claude 2026-07-31 06:33:06 +00:00
  • 4867d4f780 [CRITICAL] Enable ixformer native V1/V2 paged attention kernels Claude 2026-07-31 06:18:32 +00:00
  • 39e32343eb [ARCH] CCCL-derived paged attention kernel architecture + Triton rewrite Claude 2026-07-31 04:13:07 +00:00
  • 2316199c97 [FIX] V2 shape mismatch bug — v_padded used num_heads for kv_h tensor Claude 2026-07-31 03:52:23 +00:00
  • cd0d9e1a91 [OPT] Fix online softmax bug in Triton V2 Phase 1 Claude 2026-07-30 16:16:56 +00:00
  • d9bbef54d8 [OPT] Complete GQA broadcast — V weighted sum also avoids expansion Claude 2026-07-30 16:15:34 +00:00
  • 8951d74936 [OPT] Raise max-seq-len-to-capture to 65536 for more CUDA graph coverage dylanyunlon 2026-07-30 16:14:57 +00:00
  • 0c60ed8784 [OPT] GQA broadcast in V2 — eliminate 1GB/step memory allocation Claude 2026-07-30 16:13:54 +00:00
  • 7ad59e781f [OPT] MoE prefill: sorted-token grouped GEMM (contiguous per-expert access) dylanyunlon 2026-07-30 16:12:38 +00:00