Be more consistent in the naming convention for the various RET instructions to specify in terms of bitwidth. Helps prevent future scheduler model mismatches like those that were only addressed in D44687. Differential Revision: https://reviews.llvm.org/D113302
28 lines
631 B
YAML
28 lines
631 B
YAML
# RUN: not llc -march=x86-64 -run-pass none -o /dev/null %s 2>&1 | FileCheck %s
|
|
|
|
--- |
|
|
|
|
define void @test() {
|
|
entry:
|
|
%tmp = alloca [4168 x i8], align 4
|
|
ret void
|
|
}
|
|
|
|
...
|
|
---
|
|
name: test
|
|
tracksRegLiveness: true
|
|
frameInfo:
|
|
stackSize: 4040
|
|
stack:
|
|
- { id: 0, name: tmp, offset: -4176, size: 4168, alignment: 4 }
|
|
body: |
|
|
bb.0.entry:
|
|
$rsp = SUB64ri32 $rsp, 4040, implicit-def dead $eflags
|
|
; CHECK: [[@LINE+1]]:36: expected a 32 bit integer (the cfi offset is too large)
|
|
CFI_INSTRUCTION def_cfa_offset 123456789123456
|
|
$rsp = ADD64ri32 $rsp, 4040, implicit-def dead $eflags
|
|
RET64
|
|
...
|
|
|