llc编译wasm64文件报warning：'+neon' is not a recognized feature for this target (ignoring feature)

llc.exe -march=wasm64 -filetype=obj patch.ll -o patch.o   报如下warning：

'+neon' is not a recognized feature for this target (ignoring feature)
'+neon' is not a recognized feature for this target (ignoring feature)

具体意思是：当前目标wasm64没有 +neon这个特征，忽略这个特征

 

用文本工具打开patch.ll来查看target triple

; ModuleID = 'H:\svn\MyGame\Intermediate\Build\Android\MyGame\Shipping\link.ll'
source_filename = "llvm-link"
target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
target triple = "aarch64-none-linux-android21"

。。。 。。。

 

根本原因

NEON 是 ARM 架构特有的向量指令集（SIMD）扩展，用于加速多媒体和信号处理任务

WebAssembly（WASM） 是一种独立于 CPU 架构的虚拟指令集，其 SIMD 扩展名为 SIMD128，与 NEON 不兼容

WASM 的设计目的是跨平台兼容性，因此它不会直接支持特定 CPU 架构的扩展（如 NEON、AVX 等）

当编译时指定了 -march=wasm64，LLVM 会拒绝任何非 WASM 兼容的特征（如 +neon）

 

解决方案

方法 1：从IR文件patch.ll中移除 NEON 特征

① 在patch.ll中查找 neon ，搜索到如下内容：

attributes #0 = { nounwind sspstrong uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "min-legal-vector-width"="0" 
"no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" 
"no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="generic" "target-features"="+neon" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #1 = { argmemonly nounwind }
attributes #2 = { "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" 
"no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="generic" "target-features"="+neon" 
"unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #3 = { noinline nounwind sspstrong uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "min-legal-vector-width"="0" 
"no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" 
"no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="generic" "target-features"="+neno" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #4 = { nounwind }

② 注释掉所有与 neon 相关的行

 

方法2：修正上游编译过程

如果 patch.ll 是通过其他工具（如 clang）生成的，可能是上游编译过程中启用了 NEON

① 使用 clang 生成 IR 时禁用 NEON

clang -S -emit-llvm -mno-neon -O2 input.c -o patch.ll

② 检测所有编译选项，确保没有传递 -mfpu=neon 或类似的 NEON 相关标志

 

WASM 的 SIMD 替代方案

如果代码需要 SIMD 加速，应改用 WASM 的 simd128 特征

llc.exe -march=wasm64 -mattr=+simd128 -filetype=obj patch.ll -o patch.o