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 09:29:47 +00:00

a87f20a4ae test: just 1 tmem_store, no fence, no loop

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:29:16 +00:00

2b57f28968 test: zero 128 TMEM columns, skip fence

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:28:33 +00:00

25c9b70591 test: zero 2 TMEM columns

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:28:02 +00:00

01c4097ccc test: zero 32 TMEM columns

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:27:27 +00:00

3694f63ba4 test: re-enable full TMEM zeroing (128 columns)

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:26:53 +00:00

c3b6c3a5e6 test: minimal tmem_store debug (1 column + sentinels)

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:26:18 +00:00

f1aaa50326 test: re-enable TMEM zeroing with tmem_base debug

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:25:39 +00:00

a7f81331f8 test: skip TMEM zeroing again, alloc+dealloc only

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:25:08 +00:00

3f5dcd481e test: zero only 32 TMEM columns

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:24:31 +00:00

2b1c8ce7df test: re-enable all TMEM ops (alloc, zero, dealloc)

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:23:50 +00:00

acc7424a48 test: skip TMEM zeroing, just alloc+dealloc

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:23:12 +00:00

ca419c52f3 test: re-enable TMEM alloc + zero

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:22:19 +00:00

09e8ea5933 test: fix compile error, skip TMEM read

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:21:56 +00:00

69bbc21300 test: skip all TMEM ops, just test SMEM layout + descriptor

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:21:01 +00:00

a6c0ce51a2 test: skip MMA, just test descriptor values

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:20:14 +00:00

ea6b42e649 test_umma_qk: add descriptor debug output

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:18:47 +00:00

0f6907b001 UMMA: fix descriptor + idesc — use gau-nernst tutorial values

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:16:37 +00:00

9b458d2a6c test_umma_qk: clean rewrite, hardcoded HD=16, explicit core-matrix layout writes

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 09:15:44 +00:00

427410d94a UMMA: Rewrite fmha_umma_desc.cuh with correct K-major core-matrix layout + minimal QK GEMM test

biondizzle pushed to master at biondizzle/nvfp4-megamoe-kernel 2026-05-28 08:59:21 +00:00

68b4151d21 dump SMEM layout info