[midend] Add risc version gemmini matmul pass #492
Conversation
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
linuxlonelyeagle
suggested changes
Aug 28, 2025
| linalg.fill | ||
| ins(%1 : i8) | ||
| outs(%mem1 : memref<32x32xi8>) | ||
| linalg.matmul |
There was a problem hiding this comment.
Please add the content to be check.
| MatMulOptimize.cpp | ||
| MatMulVectorization.cpp | ||
| MatMulGemmini.cpp | ||
| MatMulParallelVectorization.cpp |
There was a problem hiding this comment.
I think we can rearrange the C++ files here. Use a dictionary sort.
| ShapedType ATy = A.getType().cast<ShapedType>(); | ||
| Type eleTy = ATy.getElementType(); | ||
| ShapedType BTy = B.getType().cast<ShapedType>(); | ||
| // ShapedType CTy = C.getType().cast<ShapedType>(); |
There was a problem hiding this comment.
Can we delete the content in the comments?
| Value B = op->getOperand(1); | ||
| Value C = op->getOperand(2); | ||
| // Get shape of input and output | ||
| ShapedType ATy = A.getType().cast<ShapedType>(); |
There was a problem hiding this comment.
use miler::cast(A.getType())
| ShapedType BTy = B.getType().cast<ShapedType>(); | ||
| // ShapedType CTy = C.getType().cast<ShapedType>(); | ||
|
|
||
| auto ctx = op->getContext(); |
There was a problem hiding this comment.
It's best not to use auto.
| auto ctx = op->getContext(); | ||
| // Some constants. | ||
| const Value c0 = | ||
| rewriter.create<arith::ConstantOp>(loc, rewriter.getIndexAttr(0)); |
There was a problem hiding this comment.
Do not use constants with Value.
| // This algorithm does not use the column A index. | ||
| // const Value aCol = rewriter.create<memref::DimOp>(loc, A, c1); | ||
| const Value bRow = rewriter.create<memref::DimOp>(loc, B, c0); | ||
| const Value bCol = rewriter.create<memref::DimOp>(loc, B, c1); |
There was a problem hiding this comment.
The same applies here.
| auto forI = rewriter.create<affine::AffineForOp> ( | ||
| loc, 0, rewriter.getIndexAttr(ATy.getDimSize(0)).getInt(), 1, std::nullopt); | ||
| // auto forI = rewriter.create<affine::AffineForOp>( | ||
| // loc, ValueRange{c0}, rewriter.getDimIdentityMap(), |
There was a problem hiding this comment.
Can we delete the content here?
| SmallVector<affine::AffineForOp, 6> band; | ||
| band.push_back(forI); | ||
| band.push_back(forJ); | ||
| band.push_back(forK); |
There was a problem hiding this comment.
SmallVector<affine::AffineForOp, 6> band = {init,}
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.
This PR aims to reproduce the methods proposed in the Exo paper to enable matrix multiplication acceleration based on Gemmini.