Skip to content

Test yb pipeline#12

Open
xuedinge233 wants to merge 2198 commits into
mainfrom
test-yb-pipeline
Open

Test yb pipeline#12
xuedinge233 wants to merge 2198 commits into
mainfrom
test-yb-pipeline

Conversation

@xuedinge233

Copy link
Copy Markdown
Contributor

The core Triton is a small number of people, and we receive many PRs (thank
you!). To help us review your code more quickly, if you are a new
contributor (less than 3 PRs merged) we ask that you complete the following
tasks and include the filled-out checklist in your PR description.

Complete the following tasks before sending your PR, and replace [ ] with
[x] to indicate you have done them.

  • I am not making a trivial change, such as fixing a typo in a comment.

  • I have written a PR description following these
    rules.

  • I have run pre-commit run --from-ref origin/main --to-ref HEAD.

  • Select one of the following.

    • I have added tests.
      • /test for lit tests
      • /unittest for C++ tests
      • /python/test for end-to-end tests
    • This PR does not need a test because FILL THIS IN.
  • Select one of the following.

    • I have not added any lit tests.
    • The lit tests I have added follow these best practices,
      including the "tests should be minimal" section. (Usually running Python code
      and using the instructions it generates is not minimal.)

hongziqi and others added 30 commits January 31, 2026 01:24
feat: update bishengir, support DLLVM_MAJOR_VERSION_21_COMPATIBLE in compile

Created-by: candyhong
Commit-by: candyhong
Merged-by: candyhong
Description: 1. update bishengir commit to 0501294d3e0f2b5d43ac49e2de4b8268db663e1c
2. support  DLLVM_MAJOR_VERSION_21_COMPATIBLE in compile with bishengir (default ON)
3. fix pybind not found, and other compile error
4. sync the gpu files with upstream commit 523a1b2
5. compile with nv, amd to fix undefined symbol
6. fix some triton-ascend python error
7. fix dyn_cast assertion llvm error

other: NPUIR need to use this Patch
<a href="https://gitcode.com/user-attachments/files/8224568/588467cb15cb444ea603ea683909020b.patch" target="_blank">588467cb15cb444ea603ea683909020b.patch</a>

See merge request: jeshd/triton-ascend!12
fix(launch): adapting the contextr variable of the signature

Created-by: candyhong
Commit-by: candyhong
Merged-by: candyhong
Description: 1.修复新版本中签名变量新增contexpr 相关bug问题:
![image.png](https://raw.gitcode.com/user-images/assets/8224568/62cb9f77-a4da-46a3-a069-73ffb1040ad6/image.png 'image.png')
2.修复tuple 对象转换问题
![image.png](https://raw.gitcode.com/user-images/assets/8224568/ac26043f-d074-4ebf-9c91-7492a676b2fc/image.png 'image.png')

See merge request: jeshd/triton-ascend!15
…sAndFoldGreedily to applyPatternsGreedily, update interface is to isa, get to cast, getWith to getIntOrFloatBitWidth, getFloat8E4M3FNType to Float8E4M3FNType::get, getFloat8E5M2Type to Float8E5M2Type::get.
…rOp, create add an argument, comment rewrite, isZeroIndex to isZeroInteger.
Co-authored-by: candyhong<1102229410@qq.com>



# message auto-generated for no-merge-commit merge:
!20 merge release/3.5.x-upgrade-candy-dev into release/3.5.x-upgrade

fix: remove outdated language/_utils.py and update root _utils.py

Created-by: candyhong
Commit-by: candyhong
Merged-by: candyhong
Description: python/triton/language/_utils.py中的函数方法 在3.4.x版本已经被合并到 python/triton/_utils.py 中,现删除python/triton/language/_utils.py 该文件(同步上游),并且对应注释2的幂次限制

See merge request: jeshd/triton-ascend!20
…nd add support for extension.parallel.

Co-authored-by: jeshd<chengmaofan@huawei.com>



# message auto-generated for no-merge-commit merge:
!19 merge release/3.5.x-upgrade-jeshd-dev-merge-5389ed7-cfc0a9d-compile-bug-fix into release/3.5.x-upgrade

fix: modify the extension op's parameter from _builder to _semantic and add support for extension.parallel.

Created-by: jeshd
Commit-by: jeshd
Merged-by: candyhong
Description: ## 描述
aux_ops,vec_ops,mem_ops里的扩展op的参数_builder修改为_semantic

## 修改原因
社区3.5.x版本对semantic进行了重构,op整合到class TritonSemantic中,jit编译时传入的参数为TritonSemantic中的实例

## checklist
<!--  [x] 表示选中 -->
- [ ] 是否通过本地IDE对代码进行静态检查
- [ ] 是否通过本地IDE对代码进行格式化处理
- [ ] 是否进行空指针校验
- [ ] 是否进行返回值校验
- [ ] 是否正确释放new/malloc申请的内存
- [ ] 是否充分考虑接口的异常场景
- [ ] 是否正确记录错误日志

See merge request: jeshd/triton-ascend!19
Co-authored-by: candyhong<1102229410@qq.com>



# message auto-generated for no-merge-commit merge:
!21 merge release/3.5.x-upgrade-candy-dev into release/3.5.x-upgrade

feat: set LLVM_MAJOR_VERSION_22_COMPATIBLE in CmakeLists.txt

Created-by: candyhong
Commit-by: candyhong
Merged-by: candyhong
Description: 需要TA设置LLVM_MAJOR_VERSION_22_COMPATIBLE宏下发到BS,避免在include阶段报错
![image.png](https://raw.gitcode.com/user-images/assets/8224568/c20ac526-3051-49ce-98d7-dd577fc821be/image.png 'image.png')

See merge request: jeshd/triton-ascend!21
Co-authored-by: candyhong<1102229410@qq.com>



# message auto-generated for no-merge-commit merge:
!22 merge release/3.5.x-upgrade-candy-dev into release/3.5.x-upgrade

fix(test): temporarily disable hang-inducing test cases

Created-by: candyhong
Commit-by: candyhong
Merged-by: candyhong
Description: 以下是卡死用例,暂时注释,避免影响批跑流程      
third_party/ascend/unittest/pytest_ut/test_asm_scalar.py
third_party/ascend/unittest/pytest_ut/test_atomic_add.py
third_party/ascend/unittest/pytest_ut/test_atomic_max.py
third_party/ascend/unittest/pytest_ut/test_atomic_min.py
third_party/ascend/unittest/pytest_ut/test_boundary_check.py
third_party/ascend/unittest/pytest_ut/test_ldst.py
third_party/ascend/unittest/pytest_ut/test_linearize_permute.py
third_party/ascend/unittest/pytest_ut/test_reduce_maximum.py
third_party/ascend/unittest/pytest_ut/test_reduce_minimum.py
third_party/ascend/unittest/pytest_ut/test_scalar_calc.py
third_party/ascend/unittest/pytest_ut/test_sum_vector.py
third_party/ascend/unittest/pytest_ut/test_swap.py

See merge request: jeshd/triton-ascend!22
…emporarily disable hang-inducing test cases.

Co-authored-by: jeshd<chengmaofan@huawei.com>



# message auto-generated for no-merge-commit merge:
!23 merge release/3.5.x-upgrade-jeshd-dev-merge-5389ed7-cfc0a9d-compile-bug-fix into release/3.5.x-upgrade

fix: replace _builder with _semantic in libdevice, modify the cast, temporarily disable hang-inducing test cases.

Created-by: jeshd
Commit-by: jeshd
Merged-by: candyhong
Description: ## 描述
把libdevice里的_builder替换为_semantic,修改cast接口,临时注释掉因改动带来的卡死用例
当前全部的卡死用例:
third_party/ascend/unittest/pytest_ut/test_asm_scalar.py
third_party/ascend/unittest/pytest_ut/test_atomic_add.py
third_party/ascend/unittest/pytest_ut/test_atomic_max.py
third_party/ascend/unittest/pytest_ut/test_atomic_min.py
third_party/ascend/unittest/pytest_ut/test_boundary_check.py
third_party/ascend/unittest/pytest_ut/test_ldst.py
third_party/ascend/unittest/pytest_ut/test_linearize_permute.py
third_party/ascend/unittest/pytest_ut/test_reduce_maximum.py
third_party/ascend/unittest/pytest_ut/test_reduce_minimum.py
third_party/ascend/unittest/pytest_ut/test_scalar_calc.py
third_party/ascend/unittest/pytest_ut/test_sum_vector.py
third_party/ascend/unittest/pytest_ut/test_swap.py
third_party/ascend/unittest/pytest_ut/test_annotations.py
third_party/ascend/unittest/pytest_ut/test_hoistbroadcast.py
third_party/ascend/unittest/pytest_ut/test_linearize.py
third_party/ascend/unittest/pytest_ut/test_reduce_min_4_keepdim_True_with_index_op.py

## checklist
<!--  [x] 表示选中 -->
- [ ] 是否通过本地IDE对代码进行静态检查
- [ ] 是否通过本地IDE对代码进行格式化处理
- [ ] 是否进行空指针校验
- [ ] 是否进行返回值校验
- [ ] 是否正确释放new/malloc申请的内存
- [ ] 是否充分考虑接口的异常场景
- [ ] 是否正确记录错误日志

See merge request: jeshd/triton-ascend!23
fix: update MAP_SIGTYPE_TO_INT for i4 and u1

Created-by: candyhong
Commit-by: candyhong
Merged-by: zhang-chunli01
Description: 参考文档补齐i4 和u1 的信号:
https://www.hiascend.com/document/detail/zh/canncommercial/850/API/appdevgapi/aclpythondevg_01_0912.html
修复以下报错:
![image.png](https://raw.gitcode.com/user-images/assets/8224568/b6a4dfa0-07ce-4ab1-9111-e77635c62776/image.png 'image.png')

See merge request: jeshd/triton-ascend!24
hipudding and others added 30 commits April 27, 2026 16:43
Use update-alternatives to create clang/clang++/lld symlinks
pointing to the -15 versions, so CMake can find them without
requiring full paths.
* ci: use CANN images in integration tests

Switch integration-tests-ascend to use pre-built CANN images
with dynamic tag based on matrix parameters. Move chip_type
to runner-preparation config (hardware-bound), and define
cann_version/os/python_version as local matrix arrays for
easy expansion.

* ci: add cann_version, os, chip_type to cache key

Include matrix dimensions (cann_version, os, chip_type) in the cache
key and restore-keys to prevent different image environments from
sharing the same cache.
Add blue yellow pipeline
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.