[Mlir-commits] [mlir] [mlir] Improve dialect conversion failure diagnostics (PR #182729)

[Jeongseok Son]

================
@@ -432,7 +432,7 @@ func.func @test_lookup_without_converter() {
 // expected-remark at -1 {{applyPartialConversion failed}}
 
 func.func @test_skip_1to1_pattern(%arg0: f32) {
-  // expected-error at +1 {{failed to legalize operation 'test.type_consumer'}}
+  // expected-error at +1 {{failed to legalize operation 'test.type_consumer' that was explicitly marked illegal: "test.type_consumer"\((%[0-9]+|<<UNKNOWN SSA VALUE>>)\) : \(f32\) -> \(\)}}
----------------
jeongseokson wrote:

Yeah I should probably have used "expected-error-re." Based on the -test-legalize-patterns configuration, I think it can vary, so I used regex for that previously.
But yes, the detail was probably not necessary to enforce through the test, so I just simplified the test. Sorry, I thought I ran the tests locally but probably didn't. It passed locally this time.

https://github.com/llvm/llvm-project/pull/182729