biondizzle

0 Followers · 0 Following

• Joined on 2025-12-10

Repositories

25

Projects Packages Public Activity Starred Repositories

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 19:32:44 +00:00

aa41cfa2e5 Multi-head FMHA kernel (Milestone 5): grid launch with MHA/MQA/batch support

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 19:18:03 +00:00

6af2feb42a TMA 5D test: element stride decomposition

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 19:12:26 +00:00

96f2f0bb90 auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 19:09:51 +00:00

015435b1ab auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 19:08:10 +00:00

41343fdc6b auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 17:00:22 +00:00

a723b524f7 TMA alignment test

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:58:32 +00:00

c54a83960d TMA debug: fix globalStrides to tensorRank-1 elements

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:55:42 +00:00

944e567b6c TMA debug: test various CUtensorMap configs

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:51:42 +00:00

55d289c65b Fix TMA: use CU_TENSOR_MAP_DATA_TYPE_BFLOAT16 not UINT16

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:46:58 +00:00

0fd3e12a52 Fix TMA test: globalStrides in bytes not elements

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:45:05 +00:00

ad8050bbad WIP: TMA load test infrastructure (manual compile needed)

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:42:26 +00:00

d9df1e6486 auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:40:52 +00:00

a4211559cf auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:39:47 +00:00

3b8fdcc823 auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:36:54 +00:00

072fbf0b5d auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:35:05 +00:00

090f2866ae Update CURRENT_ISSUE: 6-warp Milestone 1 complete

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:34:16 +00:00

b3020c2811 6-warp specialized FMHA kernel — ALL HD=16/64/128/256 PASS cos 0.999997+

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:31:02 +00:00

e71c957561 auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:29:01 +00:00

2a6d72912a auto: pre-test commit

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 16:17:50 +00:00

e74c84458c Clean up E2M1 dequant: use LUT approach (consultant recommendation)