[Mlir-commits] [mlir] [mlir][arith][transforms] Adds Truncf f32 to f4e2m1 (PR #144157)

[llvmlistbot at llvm.org]

================
@@ -322,6 +334,74 @@ struct BFloat16TruncFOpConverter : public OpRewritePattern<arith::TruncFOp> {
   }
 };
 
+struct F4E2M1ExtFOpConverter : public OpRewritePattern<arith::ExtFOp> {
----------------
Muzammiluddin-Syed-ECE wrote:

Ok so shall I add another set of patterns that perform extf and truncf using look up tables and differentiate them from the reference implementations with a different pattern benefit?

https://github.com/llvm/llvm-project/pull/144157