Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
42 commits
Select commit Hold shift + click to select a range
7cce21c
Enhance threadblock swizzle templates with default offset parameter a…
Rachmanino Dec 22, 2025
da0eea1
[Cache] Rename sparse compress cache directory
Rachmanino Dec 22, 2025
75b67b0
Temporarily exclude sink tests from non-distributed example tests in …
Rachmanino Dec 22, 2025
f6df001
[DeepEP] Move deepep benchmark to example and allow compatible with n…
Rachmanino Nov 19, 2025
da03259
[Feat] Enhance `T.st` to support intra-node store to peer's symm memory
Rachmanino Nov 20, 2025
158e98a
use strided loop to simplify get_dispatch a bit
Rachmanino Nov 20, 2025
43e6965
[Feat] Support warp reduce operators
Rachmanino Nov 21, 2025
b13fe3f
draft notify dispatch
Rachmanino Nov 21, 2025
2f65de9
rename and refactor `T.barrier/sync_blocks`
Rachmanino Nov 21, 2025
5d68610
fix prev typo
Rachmanino Nov 21, 2025
c28e0c6
[Feat] Add `get_device_tensor` function and related test
Rachmanino Nov 22, 2025
b745a76
support elect_one_sync() and add test
Rachmanino Nov 23, 2025
01b9996
draft dispatch
Rachmanino Nov 23, 2025
9a4e5e5
suupport ld, st, warp_sync, continue and add test
Rachmanino Nov 24, 2025
ea25c7f
support warp vote and add test
Rachmanino Nov 24, 2025
8333785
support device-side wait_ne
Rachmanino Nov 25, 2025
3cefc96
refactor T.wait_* and refine dispatch test logic
Rachmanino Nov 25, 2025
1cb41e7
intra-node dispatch test passed
Rachmanino Nov 25, 2025
1d7c456
draft combine
Rachmanino Nov 27, 2025
6b6b990
support massage-only debug print
Rachmanino Nov 28, 2025
449be5b
intra-node combine test passed
Rachmanino Nov 28, 2025
bc4c6d6
unify dispatch, migrate topk_idx to u64, support cached dispatch
Rachmanino Nov 28, 2025
ed2ca7b
Refactor to pre-alloc buffers and expose interface, add benchmark
Rachmanino Dec 3, 2025
32804bc
remove redundant test
Rachmanino Dec 3, 2025
2db7a38
update doc
Rachmanino Dec 5, 2025
f66691d
use int4 vectorization for dispatch
Rachmanino Dec 10, 2025
1e8ad16
use comm_stream for comm kernels
chengyupku Dec 15, 2025
05a9300
optimze dispatch perf via skipping tensor validation
Rachmanino Dec 17, 2025
072324b
add dispatch benchmark result
Rachmanino Dec 17, 2025
a36afea
make rank as an argument of the kernel
Rachmanino Dec 17, 2025
08281a6
use cuda postproc for vectorization in combine
Rachmanino Dec 18, 2025
2c1bd1f
support int4 ld/st ptx in cuda template
Rachmanino Dec 18, 2025
d23a65e
[Feat] Support auto vectorization for ld/st to optimize combine to su…
Rachmanino Dec 19, 2025
7eddf31
lint
Rachmanino Dec 19, 2025
c37575a
upd doc
Rachmanino Dec 19, 2025
71ece5e
make ci happy
Rachmanino Dec 19, 2025
47dc366
fix review issues
Rachmanino Dec 19, 2025
35433fe
fix import error
Rachmanino Dec 19, 2025
298cb04
Add DeepEP submodule and installation script for CI
Rachmanino Dec 20, 2025
5bbd6dd
fix ci bug
Rachmanino Dec 20, 2025
5f37623
[Sync] Merge mainstream TileLang TVM-FFI features into TileScale (#47)
uv-xiao Feb 6, 2026
677cdc9
Merge remote-tracking branch 'origin/main' into wt/deepep
chengyupku Feb 6, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
6 changes: 4 additions & 2 deletions .clang-tidy
Original file line number Diff line number Diff line change
@@ -1,10 +1,12 @@
---
InheritParentConfig: true
ExtraArgs: ['-v']
ExtraArgs: []
FormatStyle: file
UseColor: true
WarningsAsErrors: '*'
ExcludeHeaderFilterRegex: '^(3rdparty|tvm)/.*$'
# FIXME: Use `ExcludeHeaderFilterRegex` instead when all maintainers upgraded their `clang-tidy`
HeaderFilterRegex: '^(?!.*(?:/|^)(3rdparty|tvm)/).*'
# ExcludeHeaderFilterRegex: '^(3rdparty|tvm)/.*$'

# NOTE: there must be no spaces before the '-', so put the comma last.
Checks: >-
Expand Down
2 changes: 1 addition & 1 deletion .github/ISSUE_TEMPLATE/config.yml
Original file line number Diff line number Diff line change
@@ -1 +1 @@
blank_issues_enabled: false
blank_issues_enabled: true
63 changes: 63 additions & 0 deletions .github/ISSUE_TEMPLATE/release-plan.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,63 @@
name: "Release Plan"
description: "Plan the next release"
title: "[Release Plan] vX.Y.Z"
labels:
- release-plan
- tracking
assignees: []
body:
- type: input
id: version
attributes:
label: "Version"
placeholder: "v0.2.0"
validations:
required: true

- type: input
id: milestone
attributes:
label: "Milestone"
description: "Link or name of the milestone for this release"
placeholder: "https://github.com/tile-ai/tilelang/milestone/XX"

- type: textarea
id: scope
attributes:
label: "Scope"
description: "Goals and non-goals (brief)"
placeholder: |
- Goals: ...
- Non-goals: ...

- type: textarea
id: tasks
attributes:
label: "Tasks"
description: "Task list; link issues/PRs"
value: |
- [ ] Features
- [ ] Fixes
- [ ] Docs
- [ ] API/Breaking changes
- [ ] Benchmarks
- [ ] Release notes

- type: checkboxes
id: readiness
attributes:
label: "Readiness"
options:
- label: "All planned issues closed or deferred"
- label: "Docs updated"
- label: "CI green; artifacts verified"
- label: "Release notes drafted"

- type: textarea
id: notes
attributes:
label: "Notes"
description: "Risks or communications (optional)"
placeholder: |
- Risk: ...
- Communication: ...
8 changes: 4 additions & 4 deletions .github/workflows/amd_ci.yml
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@ jobs:
runs-on: [self-hosted, amd, gpu]

permissions:
contents: write
contents: write

steps:
- name: Checkout repository
Expand Down Expand Up @@ -56,7 +56,7 @@ jobs:
echo "------------------------------------"
exit 1
fi

- name: Commit and Push Changes
uses: stefanzweifel/git-auto-commit-action@v5
with:
Expand Down Expand Up @@ -86,7 +86,7 @@ jobs:
set -e
REQS_HASH=$(sha256sum requirements-rocm.txt | cut -d ' ' -f 1)
MARKER="${{ runner.tool_cache }}/.venv_marker_${{ env.PYTHON_VERSION }}_${REQS_HASH:0:8}"

echo "Installing requirements"
if [[ -f "$MARKER" ]] && [[ -f "${{ runner.tool_cache }}/${{ env.VENV_DIR }}/bin/activate" ]]; then
echo "venv exists and hash matches – reuse it"
Expand Down Expand Up @@ -117,4 +117,4 @@ jobs:
source "${{ runner.tool_cache }}/${{ env.VENV_DIR }}/bin/activate"
cd testing/python/amd
unset PYTHONPATH
python -m pytest -v test_tilelang_test_amd.py
python -m pytest -v test_tilelang_test_amd.py
Loading
Loading