[llvm] [WebAssembly] Added vectorized version of fexp10 to the supported list (PR #150564)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-webassembly

Author: Jasmine Tang (badumbatish)

<details>
<summary>Changes</summary>

Fixes the crash of https://github.com/llvm/llvm-project/issues/117200.

The default behavior in TargetLoweringBase is only scalar floats on fexp are supported by default, not the vectorized version. This PR adds `ISD::FEXP10` to the supported list.

---
Full diff: https://github.com/llvm/llvm-project/pull/150564.diff


2 Files Affected:

- (modified) llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp (+1-1) 
- (added) llvm/test/CodeGen/WebAssembly/libcall_vectorized.ll (+33) 


``````````diff
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
index bf2e04caa0a61..780c7028545f9 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
@@ -284,7 +284,7 @@ WebAssemblyTargetLowering::WebAssemblyTargetLowering(
 
     // Expand float operations supported for scalars but not SIMD
     for (auto Op : {ISD::FCOPYSIGN, ISD::FLOG, ISD::FLOG2, ISD::FLOG10,
-                    ISD::FEXP, ISD::FEXP2})
+                    ISD::FEXP, ISD::FEXP2, ISD::FEXP10})
       for (auto T : {MVT::v4f32, MVT::v2f64})
         setOperationAction(Op, T, Expand);
 
diff --git a/llvm/test/CodeGen/WebAssembly/libcall_vectorized.ll b/llvm/test/CodeGen/WebAssembly/libcall_vectorized.ll
new file mode 100644
index 0000000000000..2d1056fb44938
--- /dev/null
+++ b/llvm/test/CodeGen/WebAssembly/libcall_vectorized.ll
@@ -0,0 +1,33 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+
+; RUN: llc < %s -disable-wasm-fallthrough-return-opt -wasm-keep-registers  -mattr=+simd128 | FileCheck %s
+
+target triple = "wasm32-unknown-unknown"
+
+declare <4 x float> @llvm.exp10.v4f32(<4 x float>)
+
+define <4 x float> @exp10_f32v4(<4 x float> %v) {
+; CHECK-LABEL: exp10_f32v4:
+; CHECK:         .functype exp10_f32v4 (v128) -> (v128)
+; CHECK-NEXT:  # %bb.0: # %entry
+; CHECK-NEXT:    local.get $push12=, 0
+; CHECK-NEXT:    f32x4.extract_lane $push0=, $pop12, 0
+; CHECK-NEXT:    call $push1=, exp10f, $pop0
+; CHECK-NEXT:    f32x4.splat $push2=, $pop1
+; CHECK-NEXT:    local.get $push13=, 0
+; CHECK-NEXT:    f32x4.extract_lane $push3=, $pop13, 1
+; CHECK-NEXT:    call $push4=, exp10f, $pop3
+; CHECK-NEXT:    f32x4.replace_lane $push5=, $pop2, 1, $pop4
+; CHECK-NEXT:    local.get $push14=, 0
+; CHECK-NEXT:    f32x4.extract_lane $push6=, $pop14, 2
+; CHECK-NEXT:    call $push7=, exp10f, $pop6
+; CHECK-NEXT:    f32x4.replace_lane $push8=, $pop5, 2, $pop7
+; CHECK-NEXT:    local.get $push15=, 0
+; CHECK-NEXT:    f32x4.extract_lane $push9=, $pop15, 3
+; CHECK-NEXT:    call $push10=, exp10f, $pop9
+; CHECK-NEXT:    f32x4.replace_lane $push11=, $pop8, 3, $pop10
+; CHECK-NEXT:    return $pop11
+entry:
+  %r = call <4 x float> @llvm.exp10.v4f32(<4 x float> %v)
+  ret <4 x float> %r
+}

``````````

</details>


https://github.com/llvm/llvm-project/pull/150564