[Release] Bump version into 0.1.15 (#3309)

[CUDA][Reduce] Add PassConfig for FP32x2 accumulation (#3128)

[CUDA][Reduce] Add global control for FP32x2 accumulation

[Release] Bump versin into 0.1.13 (#2826)

Bump versin into 0.1.13

[Release] Bump version to 0.1.12 (#2544)

Bump version to 0.1.12

[Release] Bump version to 0.1.11 (#2354)

Bump version to 0.1.11

[Bugfix] Fix ROCm FP4 packed buffer map key (#2256)

Fix ROCm FP4 packed buffer map key

[Refactor] Strip build machine paths from LOG messages in wheel relea…

…ses (#2080)

* [Refactor] Use TVM_LOG_CUSTOMIZE to strip build paths from LOG messages in release builds

Enable TVM's custom logging hook (TVM_LOG_CUSTOMIZE) and provide our own
LogMessageImpl/LogFatalImpl that conditionally omit source file paths.
In wheel/release builds (detected via CIBUILDWHEEL or SKBUILD_STATE env
vars), LOG(WARNING) and friends no longer leak CI machine paths, showing
only the message. Local dev builds keep full paths for debugging.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* [Refactor] Downgrade noisy LOG(WARNING) to DLOG(WARNING)

These warnings about TMA/swizzle layout fallbacks and warp specialization
status are development diagnostics, not actionable for end users. Use DLOG
so they are compiled out in release (wheel) builds.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* [Chore] Apply clang-format

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

[Release] Bump version into v0.1.8 (#1853)

Bump version into 0.1.8

[Release] Bump Version into v0.1.7.post3 (#1685)

Bump Version into v0.1.7.post3

[CUDA] Enhance Broadcast Codegen for Symbolic Value (#1669)

* Enhance CUDA code generation for BroadcastNode by implementing compile-time constant folding and runtime broadcasting for various lane configurations. Improved handling for 4-bit and 8-bit integer types, ensuring correct replication and type casting in output expressions. This update increases performance and correctness in CUDA kernel generation.

* add test

* lint fix

* fix