xyyy1420 00267e43aa [BugFix] add target context and avoid redundant re-lowering in TLCPUSourceWrapper (#1899) ...

fix(cpu): skip redundant re-lowering in TLCPUSourceWrapper when mods already exist

TLCPUSourceWrapper.parse_source_information() unconditionally called
get_annotated_mod() to re-lower the IR, even when device_mod and
host_mod were already passed in from tilelang.lower(). This caused a
crash when the kernel uses T.copy (which generates T.Parallel loops),
because VectorizePlanner requires Target::Current() during lowering.

Fix: check if device_mod/host_mod already exist before re-lowering,
matching the pattern used by TLCUDASourceWrapper. Also add self.target
to the context manager for the fallback path.

Added test_matmul_with_copy_cython to reproduce the issue.

2026-03-05 13:38:05 +08:00

..

amd

[AMD] Fix gfx950 ci and add 16x16x32_bf16/fp16 instructions support (#1878)

2026-02-25 19:46:46 +08:00

analysis

[Enhancement] Update examples and tests for improved type handling functionality (#1448)

2025-12-17 14:41:25 +08:00

arith

[Refactor] Unify @jit and @lazy_jit into a single @jit decorator (#1632)

2026-01-09 15:41:39 +08:00

autotune

Refactor CUDA version checks for compute 9.0 (#1893)

2026-03-03 18:28:11 +08:00

cache

[Bugfix] Open Rocm ci test and fix some bugs. (#1443)

2026-01-06 02:18:29 +08:00

carver

Refactor CUDA version checks for compute 9.0 (#1893)

2026-03-03 18:28:11 +08:00

components

[Bugfix] Open Rocm ci test and fix some bugs. (#1443)

2026-01-06 02:18:29 +08:00

cpu

[BugFix] add target context and avoid redundant re-lowering in TLCPUSourceWrapper (#1899)

2026-03-05 13:38:05 +08:00

cuda

[CUDA][Feature] Add packed FP32x2 math intrinsics and auto vectorized support (#1839)

2026-02-12 12:50:30 +08:00

debug

[AMD] Fix gfx950 ci and add 16x16x32_bf16/fp16 instructions support (#1878)

2026-02-25 19:46:46 +08:00

fastmath

[Enhancement] Update examples and tests for improved type handling functionality (#1448)

2025-12-17 14:41:25 +08:00

issue

[BugFix] Fix Hopper TMA lowering without warp specialization (#1840)

2026-02-25 15:19:36 +08:00

jit

[Enhancement] GEMM V2 on SM90/SM100 CuTeDSL backend (#1855)

2026-02-23 01:22:47 +08:00

kernel

[CUDA] Support tcgen5mma gemm ts (#1866)

2026-02-28 16:33:13 +08:00

language

[BugFix] Fix type mismatch when lowering to AtomicAddx2 template (#1898)

2026-03-04 20:44:12 +08:00

layout

[Feature] Introduce higher-dimensional gemm layout support (#1798)

2026-02-25 15:50:58 +08:00

math

[Enhancement] Update examples and tests for improved type handling functionality (#1448)

2025-12-17 14:41:25 +08:00

metal

[Codegen] Metal codegen on Linux (#1857)

2026-02-19 11:19:17 +08:00

profiler

[Enhancement] Add dynamic symbolic constraints support for Profiler benchmarking (#1753)

2026-01-30 16:31:24 +08:00

runtime

[Enhancement] Update examples and tests for improved type handling functionality (#1448)

2025-12-17 14:41:25 +08:00

tilelibrary

[AMD] Fix ROCm FP8 dtype selection and MFMA support on gfx942/gfx950 (#1743)

2026-01-29 12:07:30 +08:00

transform

[BugFix] ArgBinder: relax shared-shape binding for unused nullable buffers (#1870)

2026-02-24 11:43:55 +08:00

utils

[Lint] Phaseout Yapf format and embrace ruff format (#1417)

2025-12-12 17:22:18 +08:00

webgpu

[Enhancement] Update examples and tests for improved type handling functionality (#1448)

2025-12-17 14:41:25 +08:00