Issues

• [packaging] deb ships unstripped libtriton.so (263 MB); rpm carries build-time artifacts — self-assigned follow-up

  flagos-ai/FlagTree#1309 · Filter by author shiptux opened on Sep 29, 2026

• [Roadmap][DRAFT] FlagTree on AMD Radeon — 26Q4 / 27Q1

  flagos-ai/FlagTree#1256 · Filter by author zhangnju opened on Sep 21, 2026
  1

• [BUG][xpu] TRITONXPU_COMPARE_FUSION=1 时 cmpf+masked store 在 tile≥4096 编译崩溃（triton_xpu.cmpf result 类型被判成 ui32）

  bugflagos2.2-rc2P1
  Bug

  flagos-ai/FlagTree#1249 · Filter by author JosephNew opened on Sep 21, 2026
  1

  

• [BUG][xpu] tl.exp2 被降级成 exp（静默错结果·比值 1/ln2），且 block≥1024 时 triton_xpu.unpack 编译崩溃

  bugflagos2.2-rc2P1
  Bug

  flagos-ai/FlagTree#1248 · Filter by author JosephNew opened on Sep 21, 2026
  3

  

• [BUG][Ascend] Lost stores: masked store skips 2xfp32 vector groups nondeterministically

  flagos-ai/FlagTree#1245 · Filter by author O1iveira7 opened on Sep 20, 2026
  1

• [BUG][Ascend] TritonToLinalgIncubated hits llvm_unreachable at TritonToLinalgPass.cpp:440 on a kernel with a dead block containing an early return

  flagos-ai/FlagTree#1244 · Filter by author O1iveira7 opened on Sep 20, 2026

• [BUG][Ascend]TritonToLinalgIncubated: unconditional cast<RankedTensorType> fails on a non-ranked-tensor type

  flagos-ai/FlagTree#1243 · Filter by author O1iveira7 opened on Sep 20, 2026

• [BUG][Ascend] TritonToLinalgIncubated leaves a live memref→tensor materialization and aborts with a dangling OpOperand assertion

  flagos-ai/FlagTree#1242 · Filter by author O1iveira7 opened on Sep 20, 2026

• [BUG][Ascend] TritonToLinalgIncubated asserts "Cannot destroy a value that still has uses!" (BlockOperand) on valid Triton kernel with nested opaque control flow

  flagos-ai/FlagTree#1241 · Filter by author O1iveira7 opened on Sep 20, 2026

• [BUG][Ascend] PlanMemory pass crashes with "buffer memory has been released"

  flagos-ai/FlagTree#1237 · Filter by author togepro opened on Sep 20, 2026

• [BUG][iluvatar] make_ttgir crash in TritonILUVATARGPUSmeLoad: tt.expand_dims layout conflict with SME mask layout

  flagos-ai/FlagTree#1235 · Filter by author O1iveira7 opened on Sep 20, 2026

• [HCU/gfx936] fp64 tl.dot silently falls back to FMA — please add upstream support for the fp64 matrix unit (v_mmac_16x16x4_f64)

  flagos-ai/FlagTree#1231 · Filter by author w1120029931-bit opened on Sep 20, 2026

• [BUILD] Ascend wheel does not declare cann-shmem, so pip install succeeds and import fails later

  flagos-ai/FlagTree#1229 · Filter by author tengqm opened on Sep 20, 2026

• [BUG][iluvatar] output corrupted when a redundant runtime branch wraps a load inside the loop

  flagos-ai/FlagTree#1228 · Filter by author O1iveira7 opened on Sep 19, 2026
  1

• [BUG][iluvatar] tl.assume inside a dead if branch makes the kernel drop its final atomic_add

  flagos-ai/FlagTree#1227 · Filter by author O1iveira7 opened on Sep 19, 2026

• [BUG][mthreads]软件流水线在循环内含运行时条件规约/store 块时静默算错（num_stages=3 触发，改 1 恢复）

  flagos-ai/FlagTree#1223 · Filter by author O1iveira7 opened on Sep 18, 2026

• [BUG][mthreads] 同一 kernel、同一输入，仅改变编译选项 num_stages 与 num_warps得到不同结果

  flagos-ai/FlagTree#1222 · Filter by author O1iveira7 opened on Sep 18, 2026

• [BUG][mthreads][ILUVATAR] 前端解析崩溃：嵌套 @triton.jit helper 接收调用方局部 tl.constexpr 时触发 region isolation（'tt.splat' op using value defined outside the region）

  flagos-ai/FlagTree#1221 · Filter by author O1iveira7 opened on Sep 18, 2026
  1

• [Ascend] _topk_topp_kernel fails to compile — ConvertTritonIRToLinalgIR PassManager::run failed

  bugflagos2.2-rc2P1
  flagos-ai/FlagTree#1220 · Filter by author Al2O3Al2O3 opened on Sep 18, 2026
  1

  

• [BUG][Ascend] Launcher workspace size overflows in 32-bit multiplication before uint64_t assignment

  flagos-ai/FlagTree#1194 · Filter by author YaooXu opened on Sep 16, 2026
  2

• [MThreads][TLE] Reusing a TME descriptor across K iterations crashes llc: cannot allocate physical SH register for TME operand

  flagos-ai/FlagTree#1183 · Filter by author Zhang-kg opened on Sep 16, 2026
  1

• Improve integer divisibility specialization to enable matmul cp.async pipelining

  flagos-ai/FlagTree#1182 · Filter by author lizhangyu258 opened on Sep 16, 2026

• [BUG] Incorrect matmul results with ​num_warps=8​: missing cross-warp-group barrier after WGMMA wait

  flagos-ai/FlagTree#1180 · Filter by author lizhangyu258 opened on Sep 16, 2026

• [Bug][metax][TLE] smem support is incomplete: scalar-indexed local_ptr asserts at LoadStoreOpToLLVM.cpp:427, smem atomics are unsupported, and smem+histogram silently corrupts data

  metax
  flagos-ai/FlagTree#1152 · Filter by author LittleShun1214 opened on Sep 11, 2026

• [iluvatar] runtime path imports pytest: triton/spec/iluvatar/triton/testing.py::nvsmi reaches triton._internal_testing

  flagos-ai/FlagTree#1142 · Filter by author tengqm opened on Sep 10, 2026