[Mlir-commits] [mlir] [mlir][linalg] Support quantized NCHW/FCHW convolution in img2col (PR #225308)

[Dmitry Manjiev]

================
@@ -1553,6 +1553,10 @@ rewriteInIm2Col(RewriterBase &rewriter,
 FailureOr<std::pair<Operation *, Operation *>>
 rewriteInIm2Col(RewriterBase &rewriter, linalg::Conv2DNchwFchwOp convOp);
 
+/// Quantized variant of rewriteInIm2Col for linalg::Conv2DNchwFchwOp.
----------------
Dima365 wrote:

Done, thanks!

https://github.com/llvm/llvm-project/pull/225308