[Mlir-commits] [mlir] [mlir][spirv] Add support for VectorAnyINTEL capability (PR #68034)

Jakub Kuderski llvmlistbot at llvm.org
Sun Oct 22 11:44:36 PDT 2023


================
@@ -18,17 +18,17 @@ func.func @expvec(%arg0 : vector<3xf16>) -> () {
 
 // -----
 
-func.func @exp(%arg0 : i32) -> () {
-  // expected-error @+1 {{op operand #0 must be 16/32/64-bit float or vector of 16/32/64-bit float values}}
-  %2 = spirv.CL.exp %arg0 : i32
+func.func @exp_any_vec(%arg0 : vector<5xf32>) -> () {
+  // CHECK: spirv.CL.exp {{%.*}} : vector<5xf32>
+  %2 = spirv.CL.exp %arg0 : vector<5xf32>
   return
 }
 
 // -----
 
-func.func @exp(%arg0 : vector<5xf32>) -> () {
-  // expected-error @+1 {{op operand #0 must be 16/32/64-bit float or vector of 16/32/64-bit float values of length 2/3/4}}
-  %2 = spirv.CL.exp %arg0 : vector<5xf32>
+func.func @exp(%arg0 : i32) -> () {
+  // expected-error @+1 {{op operand #0 must be 16/32/64-bit float or vector of 16/32/64-bit float values}}
+  %2 = spirv.CL.exp %arg0 : i32
----------------
kuhar wrote:

Same here and in the other test below

https://github.com/llvm/llvm-project/pull/68034


More information about the Mlir-commits mailing list