add support for ldexp in longvectorloweringpass (#1286)

lib/LongVectorLoweringPass.cpp

@@ -231,6 +231,7 @@ Function *getBIFScalarVersion(Function &Builtin) {
   case clspv::Builtins::kHalfRsqrt:
   case clspv::Builtins::kHalfSin:
   case clspv::Builtins::kHalfTan:
+  case clspv::Builtins::kLdexp:
   case clspv::Builtins::kLog:
   case clspv::Builtins::kLog2:
   case clspv::Builtins::kMax:

test/LongVectorLowering/ldexp.ll

@@ -0,0 +1,37 @@
+; RUN: clspv-opt --passes=long-vector-lowering %s -o %t.ll
+; RUN: FileCheck %s < %t.ll
+
+target datalayout = "e-p:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024"
+target triple = "spir-unknown-unknown"
+
+define spir_kernel void @test(ptr addrspace(1) %x, ptr addrspace(1) %k) {
+entry:
+  %0 = load <8 x float>, ptr addrspace(1) %x, align 32
+  %1 = load <8 x i32>, ptr addrspace(1) %k, align 32
+  %call = call spir_func <8 x float> @_Z5ldexpDv8_fDv8_i(<8 x float> %0, <8 x i32> %1)
+  store <8 x float> %call, ptr addrspace(1) %x, align 32
+  ret void
+}
+
+declare spir_func <8 x float> @_Z5ldexpDv8_fDv8_i(<8 x float>, <8 x i32>)
+
+; CHECK-LABEL: @test
+; CHECK-NOT: call spir_func <8 x float> @_Z5ldexpDv8_fDv8_i
+; CHECK-COUNT-8: {{.*}} = call spir_func float @_Z5ldexpfi(float {{.*}}, i32 {{.*}})
+; CHECK-NOT: call spir_func <8 x float> @_Z5ldexpDv8_fDv8_i
+
+define spir_kernel void @test2(ptr addrspace(1) %x, ptr addrspace(1) %k) {
+entry:
+  %0 = load <16 x float>, ptr addrspace(1) %x, align 32
+  %1 = load i32, ptr addrspace(1) %k, align 32
+  %call = call spir_func <16 x float> @_Z5ldexpDv16_fi(<16 x float> %0, i32 %1)
+  store <16 x float> %call, ptr addrspace(1) %x, align 32
+  ret void
+}
+
+declare spir_func <16 x float> @_Z5ldexpDv16_fi(<16 x float>, i32)
+
+; CHECK-LABEL: @test2
+; CHECK-NOT: call spir_func <16 x float> @_Z5ldexpDv8_fi
+; CHECK-COUNT-16: {{.*}} = call spir_func float @_Z5ldexpfi(float {{.*}}, i32 {{.*}})
+; CHECK-NOT: call spir_func <16 x float> @_Z5ldexpDv8_fi