COMMITS
August 23, 2026
L
[Runtime][Cache] Remove separate cache temporary directory (#3069)
Lei Wang committed
August 22, 2026
L
[Build] Support Cython 3.3 with the Python 3.9 limited API (#3068)
Lei Wang committed
C
[CUDA] Fix warp shuffle for half, bfloat16, and FP8 (#3056)
Chennes committed
August 21, 2026
L
[Layout][Config] Add environment override for layout cost model (#3061)
Lei Wang committed
L
[CUDA][Reduce] Add per-reduce control for FP32x2 accumulation (#3057)
Lei Wang committed
August 20, 2026
C
[Skill] Add TileLang semantic validation skill (#3054)
Chaofan Lin committed
Y
[CUDA] Remove `T.{reads,writes}` for `T.tma_{gather4,scatter4}` (#3053)
Yongqi Zhuo committed
L
[Layout][Inference] Restore register-count as the default cost model (#3055)
Lei Wang committed
S
[Bugfix] Carry memory_order through vectorized atomic_add (#2924)
Srijan Keshri committed
P
[Misc] Remove incorrect ASF license headers from src files (#3051)
penguin_wwy committed
August 19, 2026
P
[CodeGen] Emit #line directives from TIR spans (#3048)
penguin_wwy committed
L
[TIR][Transform] Handle grid barriers and unbounded pointer ranges (#3050)
Lei Wang committed
August 18, 2026
L
[BugFix][Layout] Avoid thread-indexed wide reducer finalize readback (#3049)
Lei Wang committed
C
[CUDA] Fix FP8 min/max codegen (#3047)
Chennes committed
August 17, 2026
L
[Lang][Reducer] Allow conditional reducer finalization (#3044)
Lei Wang committed
P
[TIR][Transform] Preserve source spans across lowering passes (#2966)
penguin_wwy committed
L
C
[Analysis] Reject `T.Parallel` indexing of local buffers (#3041)
Chaofan Lin committed
C
[CUDA] Add __ldg overloads for 16-bit CUTLASS types (#3028)
Chennes committed
R
August 16, 2026
L
[Layout][Inference] Add IO-aware cost model for free-mode selection (#2960)
Lei Wang committed
August 15, 2026
L
[Lang][TIR] Make region bridge a builtin intrinsic (#2983)
Lei Wang committed
C
[BugFix] Isolate NVRTC kernel handles per adapter (#2950)
Chaofan Lin committed
August 14, 2026
G
[BugFix][Metal] Preserve pointer address spaces for byte offsets (#2925)
Gengyuan Bai committed
August 13, 2026
L
[CI][Examples] Remove TopK example from performance regression (#2958)
Lei Wang committed
L
Y
[Testing] Pin the folded-base descriptor form for static ts slices (#2951)
Yongqi Zhuo committed
Y
[BugFix][CUDA] Advance tcgen05 ld/st segment pointers in b32 columns (#2952)
Yongqi Zhuo committed
Y
[Layout][CUDA] Support reinterpreting (dtype-changing) T.view aliases (#2953)
Yongqi Zhuo committed