v2.4.1 Pack LayerNorm

What's Changed

• [Nsight] Add nsys/ncu usage, ptx/sass by @DefTruth in #44
• [DotProd][FP16] support f16x8_pack kernel by @DefTruth in #45
• [LayerNorm][FP16] Add pack support for f16x8 LD/ST by @DefTruth in #46

Full Changelog: v2.4...v2.4.1

v2.4 Pack Reduce LDST

What's Changed

• [Reduce][Kernel] Pack f16/bf16x8 & fp8/i8x16 LD/ST by @DefTruth in #43

Full Changelog: v2.3.1...v2.4

v2.3.1 f16x8 Pack Elementwise

What's Changed

• [FA2][Half] Add FA2 f16_mma_m16n8k16 kernel by @DefTruth in #35
• [Refactor][7/N] CUDA Learn Notes refactor Part-7 by @DefTruth in #36
• Clamped input range in Sigmoid kernel to prevent overflow by @Phoenix8215 in #37
• [Sigmoid][F16] Add f16x8_pack kernel, boost 1.5x ~ by @DefTruth in #39
• [Elementwise][Half] support f16x8_pack kernel, boost 1.1x by @DefTruth in #40
• [FlashAttention] replace FLOAT4 with LDST128BITS macro by @DefTruth in #41
• [RELU][FP16] Add f16x8_pack kernel, boost 2.1x by @DefTruth in #42

New Contributors

• @Phoenix8215 made their first contribution in #37

Full Changelog: v2.3...v2.3.1

v2.3 Refactor 6/N

What's Changed

• [Refactor][6/N] CUDA Learn Notes refactor Part-6 by @DefTruth in #17
• [Refactor][5/N] CUDA Learn Notes refactor Part-6 by @DefTruth in #18
• [LayerNorm][Half] support fp16x8 packed LayerNorm by @DefTruth in #19
• [Reduce][Half] add HALF2 & BFLOAT2 macro by @DefTruth in #21
• [RMSNorm][Half] support fp16x8 packed RMSNorm by @DefTruth in #22
• [Bugfix][Kernel] fixed some kernel blocks calculate errors by @DefTruth in #23
• [Elementwise][Half] support fp16x8 packed Elementwise by @DefTruth in #24
• [Elementwise][Half] support fp16x8 packed Elementwise by @DefTruth in #25
• [RELU][Half] support fp16x8 RELU kernel by @DefTruth in #26
• [RMSNorm] support f16x8_f32 RMSNorm by @DefTruth in #28
• [RMSNorm][Kernel] Add FLOAT2/HALF2_VARIANCE macro by @DefTruth in #29
• [LayerNorm][Kernel] Add HALF2 SUM/SUB/VAR macro by @DefTruth in #30
• [HGEMM] Add slicked_k&t_8x8_sliced_k_f16x4 by @DefTruth in #31
• [HGEMV][Half] support hgemv k32/k128/f16 by @DefTruth in #32
• [FlashAttention] Refactor flash_attn_1_fwd_f32 kernel by @DefTruth in #33
• Bump up to v2.3 by @DefTruth in #34

Full Changelog: v2.2...v2.3

v2.2 Refactor 5/N

What's Changed

• [Refactor][5/N] CUDA Learn Notes refactor Part-5 by @DefTruth in #15
• Bump up to v2.2 by @DefTruth in #16

Full Changelog: 2.1...v2.2

v2.1 Refactor 4/N Part-4

What's Changed

• [Refactor][4/N] CUDA Learn Notes refactor Part-4 by @DefTruth in #10
• [Refactor][4/N] CUDA Learn Notes refactor Part-4 by @DefTruth in #11
• [Refactor][4/N] CUDA Learn Notes refactor Part-4 by @DefTruth in #12
• [Refactor][4/N] CUDA Learn Notes refactor Part-4 by @DefTruth in #13
• [Refactor][4/N] CUDA Learn Notes refactor Part-4 by @DefTruth in #14

Full Changelog: v2.0...2.1

v2.0 Refactor 4/N

Full Changelog: v0.8...v2.0

v0.8

What's Changed

• Bump up to v0.8 by @DefTruth in #9

Full Changelog: v0.7...v0.8

CUDA Learn Note v0.7

Full Changelog: v0.5...v0.6

What's Changed

• Bump up to v0.7 by @DefTruth in #8

New Contributors

• @DefTruth made their first contribution in #8

Full Changelog: v0.6...v0.7

CUDA Learn Notes v0.5

Full Changelog: v0.3...v0.5