[CINN] Add ElinimateCommonFactorOfLocalIndex pass in OptimizeExprGPU
#62207
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
你的PR提交成功，感谢你对开源项目的贡献! |
tc20042008
previously approved these changes
Feb 29, 2024
zhhsplendid
reviewed
Feb 29, 2024
| } | ||
| } | ||
|
|
||
| int ExtractNumberFromExpr(const ir::Expr& expr) { |
There was a problem hiding this comment.
It seems some size, index can be int64, is int enough for cases here?
There was a problem hiding this comment.
This function only handles the factor of ir::Expr, and the ir::Expr is the index of the local variable, so we will rarely encounter int64 cases.
| void operator()(ir::Expr* expr) { ir::IRMutator<>::Visit(expr, expr); } | ||
|
|
||
| const std::unordered_map<std::string, std::vector<std::vector<ir::Expr>>>& | ||
| local_var_to_indexes() { |
There was a problem hiding this comment.
This function doesn't change the member, should we also mark a suffix const?
| "should at least load and store once."; | ||
| for (std::size_t i = 1; i < indexes.size(); ++i) { | ||
| CHECK_EQ(indexes[0].size(), indexes[i].size()) | ||
| << "We should guarantee all index vector have the same size."; |
tc20042008
reviewed
Mar 4, 2024
| namespace cinn::utils { | ||
|
|
||
| static const std::unordered_set<std::string> | ||
| kProhibitScheduleExternalFuncNames = { |
tc20042008
reviewed
Mar 4, 2024
|
|
||
| const std::unordered_set<std::string>& GetProhibitScheduleExternalFuncNames() { | ||
| static const std::unordered_set<std::string> | ||
| kProhibitScheduleExternalFuncNames = { |
tc20042008
previously approved these changes
Mar 4, 2024
…teCommonFactorOfLocalIndex
tc20042008
approved these changes
Mar 5, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
New features
PR changes
Others
Description
pcard-76996
问题背景
对 for 循环做 Reorder 操作时，某些局部变量的读取顺序并非连续的，根据下标索引计算局部变量的空间大小时，部分下标没有充分化简，导致分析出的空间变大，本 PR 通过提供
ElinimateCommonFactorOfLocalIndex解决此问题LocalAxisVisitor 之后得到的表达式：
由于
(32 * k)未充分化简，生成的 kernel 代码会变成：解决思路
从 Tensor 的视角出发：
最终效果