Issues

• Can the special clause in the LICENSE file be removed now that we're past the specified period's end date?

  tile-ai/tilelang#3371 · Filter by author mbrukman opened on Oct 1, 2026

• [BUG][CUDA] FP16/BF16 vector comparisons are scalarized instead of using packed comparisons

  bug
  tile-ai/tilelang#3302 · Filter by author sepcnt opened on Sep 29, 2026

• [BUG][Batch] 49 reproducible correctness bugs in TileLang (with runnable repros attached)

  tile-ai/tilelang#3301 · Filter by author Hughshine opened on Sep 29, 2026
  9

• [Feature Request] Support for apache-tvm-ffi 0.1.13

  enhancement
  tile-ai/tilelang#3298 · Filter by author wzhao18 opened on Sep 28, 2026
  2

  

• [BUG] Bulk TMA cache hints fail to compile and shared-memory over-alignment reduces occupancy

  tile-ai/tilelang#3295 · Filter by author sepcnt opened on Sep 28, 2026

• [BUG] [METAL] T.copy at default coalesced_width incorrectly strides global-to-shared copies

  bug
  tile-ai/tilelang#3292 · Filter by author fizzAI opened on Sep 28, 2026
  1

• [BUG][CUDA] Warp-specialized pipeline releases a stage before a non-TMA shared copy is consumed (wrong results with a register A operand)

  tile-ai/tilelang#3287 · Filter by author ghostrider0470 opened on Sep 27, 2026

• [Performance][CUDA] Late SIMT im2col lowering misses scheduling and vectorization opportunities

  tile-ai/tilelang#3274 · Filter by author sepcnt opened on Sep 24, 2026

• [Feature Request][Perf] Range contracts and optimization diagnostics for T.dynamic

  tile-ai/tilelang#3271 · Filter by author sepcnt opened on Sep 23, 2026

• [BUG] NVRTC wrapper omits host-computed scalar kernel arguments

  tile-ai/tilelang#3263 · Filter by author sepcnt opened on Sep 21, 2026
  3

• [Feature Request] [Perf] LowerMagicDiv: host-precomputed magic division for launch-invariant dynamic divisors

  enhancement
  tile-ai/tilelang#3261 · Filter by author penguin-wwy opened on Sep 21, 2026

• [BUG][CuTeDSL][FP4/FP8] DeepSeek V4 act quant: FP4 conversion/store broken against CUTLASS DSL 4.7, FP8 store rejected by libNVVM

  tile-ai/tilelang#3240 · Filter by author 0z5a opened on Sep 17, 2026
  1

• [BUG] Unsigned condition analysis constructs a negative unsigned coefficient

  bug
  tile-ai/tilelang#3235 · Filter by author sepcnt opened on Sep 15, 2026

• [BUG] Nested data-dependent serial loops inside T.Parallel fail during LowerTileOp

  bug
  tile-ai/tilelang#3225 · Filter by author sepcnt opened on Sep 14, 2026

• [BUG] T.gemm silently downgrades WGMMA to mma.sync for FP8 operands on SM90

  bug
  tile-ai/tilelang#3223 · Filter by author xuebozhang525-alt opened on Sep 14, 2026
  1

• [Feature Request] [Metal] MLX adapter for Metal kernels (vllm-metal)

  enhancement
  tile-ai/tilelang#3222 · Filter by author WindChimeRan opened on Sep 13, 2026

• [Transform][Arith] ConstrSet::Populate ordering is load-bearing in two incompatible directions (eager Analyzer::Bind snapshots)

  tile-ai/tilelang#3220 · Filter by author LeiWang1999 opened on Sep 13, 2026

• [Maint] layout_inference golden snapshots are stale on main; harness has no CI hook

  tile-ai/tilelang#3188 · Filter by author LeiWang1999 opened on Sep 9, 2026

• [Feature Request] [RFC] Add explicit CUDA cache policies for normal global-memory copies

  enhancement
  tile-ai/tilelang#3178 · Filter by author yanglf1121 opened on Sep 7, 2026
  

• [Question][Carver] TensorCorePolicy: tiles whose smem_cost × pipeline_stage exceeds smem_cap are pruned rather than downgraded, and the shared.dyn fallback in _assign_block_size looks unreachable

  question
  tile-ai/tilelang#3164 · Filter by author mocusez opened on Sep 3, 2026
  1

• [Perf] Lower T.any_of / T.all_of via warp vote intrinsics (__any_sync / __all_sync) where safe

  enhancementgood first issue
  tile-ai/tilelang#3120 · Filter by author LeiWang1999 opened on Aug 31, 2026
  2

  

• [BUG][Fuzzer][ice-on-valid-code] 8-bit-operand sparse T.gemm_sp with K > block_K crashes ThreadSync (Cannot match type int32 vs handle) instead of compiling

  tile-ai/tilelang#3037 · Filter by author Hughshine opened on Aug 17, 2026
  1

• [BUG][Fuzzer][ice-on-invalid-code] T.rng_rand/T.rng_rand_float with no preceding T.rng_init emit curand…(&) and crash nvcc, instead of a clean "init required" error

  tile-ai/tilelang#3034 · Filter by author Hughshine opened on Aug 17, 2026
  3

• [BUG][Fuzzer][ice-on-invalid-code] T.gemm on a narrow dtype (int8/fp8/bfloat16) with any sub-atom block_K crashes nvcc instead of cleanly rejecting

  tile-ai/tilelang#3032 · Filter by author Hughshine opened on Aug 17, 2026

• [BUG][Fuzzer][ice-on-valid-code] Binding T.rng_init()'s documented state handle to a variable emits void state = ; and nvcc rejects the kernel

  tile-ai/tilelang#3029 · Filter by author Hughshine opened on Aug 17, 2026
  2