Files
clang-p2996/llvm/test/CodeGen/MIR/X86/variable-sized-stack-objects.mir
Fangrui Song 40a4cbb0f2 [MIR,test] Change llc -march=x86-64 to -mtriple=x86_64
Similar to 806761a762

-mtriple= specifies the full target triple while -march= merely sets the
architecture part of the default target triple (e.g. Windows, macOS).

Therefore, -march= is error-prone and not recommended for tests without
a target triple. The issue has been benign as these MIR tests do not
utilize object file format specific detail, but it's good to change
these tests to neighbor files that use -mtriple=x86_64
2024-12-15 11:23:08 -08:00

45 lines
1.4 KiB
YAML

# RUN: llc -mtriple=x86_64 -run-pass none -o - %s | FileCheck %s
# This test ensures that the MIR parser parses variable sized stack objects
# correctly.
--- |
define i32 @test(i32 %a) {
entry:
%b = alloca i32
%x = alloca i64
%y = alloca i32, i32 %a
store i32 %a, ptr %b
store i64 2, ptr %x
%c = load i32, ptr %b
ret i32 %c
}
...
---
name: test
frameInfo:
stackSize: 24
offsetAdjustment: -16
maxAlignment: 8
adjustsStack: true
# CHECK: stack:
# CHECK-NEXT: - { id: 0, name: '', type: default, offset: -20, size: 4, alignment: 4,
# CHECK-NEXT: stack-id: default, callee-saved-register: '', callee-saved-restored: true,
# CHECK-NEXT: debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
# CHECK-NEXT: - { id: 1, name: '', type: default, offset: -32, size: 8, alignment: 8,
# CHECK-NEXT: stack-id: default, callee-saved-register: '', callee-saved-restored: true,
# CHECK-NEXT: debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
# CHECK-NEXT: - { id: 2, name: y, type: variable-sized, offset: -32, alignment: 1,
stack:
- { id: 0, offset: -20, size: 4, alignment: 4 }
- { id: 1, offset: -32, size: 8, alignment: 8 }
- { id: 2, name: y, type: variable-sized, offset: -32, alignment: 1 }
body: |
bb.0.entry:
MOV32mr $rsp, 1, _, -4, _, $edi
MOV64mi32 $rsp, 1, _, -16, _, 2
$eax = MOV32rm $rsp, 1, _, -4, _
RET64 $eax
...