Lower cxx to the LLVM dialect

[robertoraggi]

Barely enough to get started to lower cxx to the MLIR LLVM dialect.

cat x.c

int main() { return 123; }

cxx -toolchain macos -emit-ir x.c

module @x.c {
  llvm.func @main() -> i32 {
    %0 = llvm.mlir.constant(1 : i64) : i64
    %1 = llvm.alloca %0 x i32 : (i64) -> !llvm.ptr
    %2 = llvm.mlir.constant(123 : i64) : i32
    llvm.store %2, %1 : i32, !llvm.ptr
    llvm.br ^bb1(%1 : !llvm.ptr)
  ^bb1(%3: !llvm.ptr):  // pred: ^bb0
    %4 = llvm.load %1 : !llvm.ptr -> i32
    llvm.return %4 : i32
  }
}

cxx -toolchain macos -emit-ir x.c | mlir-translate --mlir-to-llvmir

; ModuleID = 'LLVMDialectModule'
source_filename = "LLVMDialectModule"

define i32 @main() {
  %1 = alloca i32, i64 1, align 4
  store i32 123, ptr %1, align 4
  br label %2

2:                                                ; preds = %0
  %3 = phi ptr [ %1, %0 ]
  %4 = load i32, ptr %1, align 4
  ret i32 %4
}

!llvm.module.flags = !{!0}

!0 = !{i32 2, !"Debug Info Version", i32 3}

cxx -toolchain macos -emit-ir x.c | mlir-translate --mlir-to-llvmir | llc

        .build_version macos, 15, 0
        .section        __TEXT,__text,regular,pure_instructions
        .globl  _main                           ; -- Begin function main
        .p2align        2
_main:                                  ; @main
        .cfi_startproc
; %bb.0:
        sub     sp, sp, #16
        .cfi_def_cfa_offset 16
        mov     w8, #123                        ; =0x7b
        mov     w0, #123                        ; =0x7b
        str     w8, [sp, #12]
        add     sp, sp, #16
        ret
        .cfi_endproc
                                        ; -- End function
.subsections_via_symbols

cxx -toolchain macos -emit-ir x.c | mlir-translate --mlir-to-llvmir | lli
echo $?

123