20100a7254
This PR merges the `triton-mlir` branch, in which we have been quietly rewriting the Triton backend from scratch to increase maintainability, stability and ultimately performance. Changes to the runtime are minimal, and this new version aims to remain backward-compatible with the previous commit. The legacy backend is now officially deprecated, but can still be accessed via the `legacy-backend` tag. Co-authored-by: Keren Zhou <kerenzhou@openai.com> Co-authored-by: Yan Chunwei <yanchunwei@outlook.com> Co-authored-by: goostavz <109190422+goostavz@users.noreply.github.com> Co-authored-by: Shintaro Iwasaki <siwasaki@fb.com> Co-authored-by: Yan Da <dyanab@connect.ust.hk> Co-authored-by: Jun Yang <yangjunpro@gmail.com> Co-authored-by: Ian Bearman <ianb@microsoft.com> Co-authored-by: Jason Ansel <jansel@jansel.net> Co-authored-by: Qingyi Liu <qingyil@nvidia.com> Co-authored-by: ben-zhang-609 <110140741+ben-zhang-609@users.noreply.github.com> Co-authored-by: Chenggang Zhao <lyricz@yeah.net> Co-authored-by: ben-zhang-609 <benzh609@gmail.com> Co-authored-by: dongdongl <dongdongl@nvidia.com>
58 lines
1.6 KiB
Plaintext
58 lines
1.6 KiB
Plaintext
# These owners will be the default owners for everything in
|
|
# the repo. Unless a later match takes precedence,
|
|
# @global-owner1 and @global-owner2 will be requested for
|
|
# review when someone opens a pull request.
|
|
* @ptillet
|
|
|
|
# --------
|
|
# Analyses
|
|
# --------
|
|
# Alias analysis
|
|
include/triton/Analysis/Alias.h @Jokeren
|
|
lib/Analysis/Alias.cpp @Jokeren
|
|
# Allocation analysis
|
|
include/triton/Analysis/Allocation.h @Jokeren
|
|
lib/Analysis/Allocation.cpp @Jokeren
|
|
# Membar analysis
|
|
include/triton/Analysis/Membar.h @Jokeren
|
|
lib/Analysis/Membar.cpp @Jokeren
|
|
# AxisInfo analysis
|
|
include/triton/Analysis/AxisInfo.h @ptillet
|
|
lib/Analysis/AxisInfo.cpp @ptillet
|
|
# Utilities
|
|
include/triton/Analysis/Utility.h @Jokeren
|
|
lib/Analysis/Utility.cpp @Jokeren
|
|
|
|
# ----------
|
|
# Dialects
|
|
# ----------
|
|
# Pipeline pass
|
|
lib/Dialect/TritonGPU/Transforms/Pipeline.cpp @daadaada
|
|
# Prefetch pass
|
|
lib/Dialect/TritonGPU/Transforms/Prefetch.cpp @daadaada
|
|
# Coalesce pass
|
|
lib/Dialect/TritonGPU/Transforms/Coalesce.cpp @ptillet
|
|
# Layout simplification pass
|
|
lib/Dialect/TritonGPU/Transforms/Combine.cpp @ptillet
|
|
|
|
# -----------
|
|
# Conversions
|
|
# -----------
|
|
# TritonGPUToLLVM
|
|
include/triton/Conversion/TritonGPUToLLVM/ @goostavz @Superjomn
|
|
lib/Conversions/TritonGPUToLLVM @goostavz @Superjomn
|
|
# TritonToTritonGPU
|
|
include/triton/Conversion/TritonToTritonGPU/ @daadaada
|
|
lib/Dialect/TritonGPU/Transforms/TritonGPUConversion.cpp @daadaada
|
|
|
|
|
|
# -------
|
|
# Targets
|
|
# -------
|
|
# LLVMIR
|
|
include/triton/Target/LLVMIR/ @goostavz @Superjomn
|
|
lib/Target/LLVMIR @goostavz @Superjomn
|
|
# PTX
|
|
include/triton/Target/PTX/ @goostavz @Superjomn
|
|
lib/Target/PTX @goostavz @Superjomn
|