clear residual op in vpto emit#855
Merged
zhangstevenunity merged 1 commit intoJul 3, 2026
Merged
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request introduces folding for pto.get_validshape intrinsics into materialized tile handles, along with erasing unused pto.set_validshape operations and performing dead code elimination (DCE) on leftover tile-handle view and allocation operations. The review feedback suggests optimizing the iterative DCE implementation to a worklist-based approach to improve complexity from UnrealizedConversionCastOp has exactly one result before accessing it.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
Comment on lines
+680
to
+699
| bool tileDceChanged = true; | ||
| while (tileDceChanged) { | ||
| tileDceChanged = false; | ||
| SmallVector<Operation *, 8> deadTileOps; | ||
| func.walk([&](Operation *op) { | ||
| if (!op->use_empty()) | ||
| return; | ||
| if (isa<pto::TReshapeOp, pto::MaterializeTileOp, pto::AllocTileOp>(op)) | ||
| deadTileOps.push_back(op); | ||
| else if (auto castOp = dyn_cast<UnrealizedConversionCastOp>(op)) { | ||
| if (castOp.getNumOperands() == 1 && | ||
| isa<pto::TileBufType>(castOp.getResult(0).getType())) | ||
| deadTileOps.push_back(op); | ||
| } | ||
| }); | ||
| for (auto *op : llvm::reverse(deadTileOps)) { | ||
| op->erase(); | ||
| tileDceChanged = true; | ||
| } | ||
| } |
Contributor
There was a problem hiding this comment.
The current iterative DCE implementation performs a full function walk (func.walk) inside a while loop for every level of dead tile operations. For a chain of
We can optimize this to UnrealizedConversionCastOp has exactly 1 result before accessing getResult(0) to prevent potential out-of-bounds assertions.
SmallVector<Operation *, 8> worklist;
func.walk([&](Operation *op) {
if (!op->use_empty())
return;
if (isa<pto::TReshapeOp, pto::MaterializeTileOp, pto::AllocTileOp>(op)) {
worklist.push_back(op);
} else if (auto castOp = dyn_cast<UnrealizedConversionCastOp>(op)) {
if (castOp.getNumOperands() == 1 && castOp.getNumResults() == 1 &&
isa<pto::TileBufType>(castOp.getResult(0).getType()))
worklist.push_back(op);
}
});
while (!worklist.empty()) {
Operation *op = worklist.pop_back_val();
SmallVector<Value, 4> operands(op->getOperands());
op->erase();
for (Value operand : operands) {
if (auto *defOp = operand.getDefiningOp()) {
if (!defOp->use_empty())
continue;
if (isa<pto::TReshapeOp, pto::MaterializeTileOp, pto::AllocTileOp>(defOp)) {
worklist.push_back(defOp);
} else if (auto castOp = dyn_cast<UnrealizedConversionCastOp>(defOp)) {
if (castOp.getNumOperands() == 1 && castOp.getNumResults() == 1 &&
isa<pto::TileBufType>(castOp.getResult(0).getType()))
worklist.push_back(defOp);
}
}
}
}
Likai-19
force-pushed
the
vpto_clear_residual_op
branch
from
6c34f51 to
eb6e8d3
Compare
Codex Review该评论由 review 机器人自动更新。
SummaryReview failed at stage Findings未生成结构化 findings，因为 review 过程提前失败。 Log Tail |
mouliangyu
approved these changes
Jun 29, 2026
Likai-19
force-pushed
the
vpto_clear_residual_op
branch
from
eb6e8d3 to
b8cc380
Compare
Likai-19
force-pushed
the
vpto_clear_residual_op
branch
from
b8cc380 to
8a9ab88
Compare
A3 板测完成（有跳过）
|
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
问题总结
VPTO lowering pipeline 完成后，module 中残留的 tile 元数据 op 阻塞了 LLVM 导出。这些 op 分为两类：
语义已完成但未清理的死 op：
pto.set_valid_shape（始终无 user）、无 user 的pto.get_valid_shape、无 user 的pto.treshape、无 user 的pto.alloc_tile、无 user 的UnrealizedConversionCastOp。可被折叠的 cast 链：
pto.get_valid_shape→UnrealizedConversionCastOp→ LLVM i64 的 cast 链，其等价的 LLVM i64 值已存在于原始pto.alloc_tile的 valid row/col 属性中，但未被复用。