Skip to content

[CINN] [New Hardware Update]Clean obsolete HIP Reduce templates #72142

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Apr 14, 2025

Conversation

feixi21
Copy link
Contributor

@feixi21 feixi21 commented Apr 9, 2025

PR Category

CINN

PR Types

Deprecations

Description

清理Hip Reduce模版函数,并简化Reduce模版函数命名

Copy link

paddle-bot bot commented Apr 9, 2025

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@paddle-bot paddle-bot bot added the contributor External developers label Apr 9, 2025
@@ -48,7 +48,6 @@ ir::Module CreateSwitchWithBroadcastConditionModule(
const auto &symbolic_arg_define = [&]() -> std::vector<ir::Expr> {
std::vector<ir::Expr> arg_defs;
for (const auto &item : symbolic_shape_var_index) {
#ifdef CINN_WITH_CUDA
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

call Node 不需要区分CUDA runtime::intrinsic::get_value_in_cuda_kernel_args和HIP/SYCL的区别吗?

@@ -366,7 +353,6 @@ void detail::CollectBucketStrategyHostFunctionVisitor::ProcessArgs(
const std::vector<ir::Argument> &args = func->args;
for (int i = 0; i < args.size(); ++i) {
if (args[i].is_var()) {
#ifdef CINN_WITH_CUDA
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

同上

@@ -68,27 +68,11 @@ class CodeGenSwitchHost : public CodeGenInvokeModule {
: CodeGenInvokeModule(m, b, vars) {}
// only support call of args get function and inner case host function call
llvm::Value *Visit(const ir::Call *op) override {
return common::DefaultDeviceTarget().arch.Match(
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

同上

@phlrain phlrain merged commit 6d3f14b into PaddlePaddle:develop Apr 14, 2025
32 checks passed
YqGe585 pushed a commit to YqGe585/Paddle that referenced this pull request May 7, 2025
…lePaddle#72142)

* remove hip call_get_value_in_kernel_args

* fix hip reduce

* remove cinn_get_value_in_hip_kernel

fix bugs
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
contributor External developers
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants