Triton 编译器发布: Triton 3.7.0 Release Notes

来源摘要

## Table of Contents - [Dialect & Frontend](#dialect--frontend) - [Backend & Compiler](#backend--compiler) - [AMD/HIP Backend](#amdhip-backend) - [NVIDIA Backend](#nvidia-backend) - [Gluon & Layout Improvements](#gluon--layout-improvements) - [Kernels & Benchmarks](#kernels--benchmarks) - [Proton Profiling](#proton-profiling) - [Testing & CI](#testing--ci) - [Build & Infrastructure](#build--infrastructure) - [Documentation](#documentation) - [Breaking Changes](#breaking-changes) - [Contributors](#contributors) --- ## Dialect & Frontend ### New Features - **`tl.squeeze` / `tl.unsqueeze`**: Added `tl.squeeze` and `tl.unsqueeze` operations to the standard library (#8924) - **Scaled BMM**: Added support for scaled batched matmul in the frontend (#9000) - **FP8 Constants**: Frontend can now create FP8 constants directly (#8882) - **Returning Constexpr from JIT**: Functions can return `constexpr` values from JIT-compiled code (#8785) - **`get_int_attr` for Out-of-Tree Walk**: Added `get_int_attr` to `Operation` to support out-of-tree IR walks (#8892) - **Optional Device Arg to `preload`**: Added optional device argument to `preload` and guardrails for cross-target preload (#8951, #8952, #9234) - **`tl.cat(can_reorder=False)`**: Added a non-reordering variant of `tl.cat` with broadcast support (#9312, #9163) - **Round f32→tf32 in Descriptor**: Added option to round f32 to tf32 inside tensor descriptors (#9295) - **Plugin Hooks & Out-of-Tree Dialects**: Added support for out-of-tree TTIR/TTGIR passes and Triton Dialect Plugins, with example documentation (#8401, #8523, #8815) ### Bug Fixes - **`desc.shape` for FP4 Padded**: Fixed `desc.shape` values for fp4-padded tensor descriptors (#9012) - **Setting Attr on Constexpr Argument**: Fixed setting attributes on constexpr argument