X86: loosen an overly aggressive MachO assertion

We would assert that the FP setup CFI used esp/rsp always.  This held up in
practice when the code was generated from IR.  However, with the integrated
assembler, it is possible to have the input be user specified assembly.  In such
a case, we cannot assume that the function implementation has a compact unwind
representation.  Loosen the assertion into a check and bail if we cannot
represent the frame pointer in the compact unwinding.

Addresses PR30453!

llvm-svn: 281986
This commit is contained in:
Saleem Abdulrasool 2016-09-20 17:05:04 +00:00
parent c8e7c98b66
commit 03ffa797ad
2 changed files with 17 additions and 2 deletions

View File

@ -546,8 +546,12 @@ protected:
// .cfi_def_cfa_register %rbp
//
HasFP = true;
assert(MRI.getLLVMRegNum(Inst.getRegister(), true) ==
(Is64Bit ? X86::RBP : X86::EBP) && "Invalid frame pointer!");
// If the frame pointer is other than esp/rsp, we do not have a way to
// generate a compact unwinding representation, so bail out.
if (MRI.getLLVMRegNum(Inst.getRegister(), true) !=
(Is64Bit ? X86::RBP : X86::EBP))
return 0;
// Reset the counts.
memset(SavedRegs, 0, sizeof(SavedRegs));

View File

@ -0,0 +1,11 @@
// RUN: llvm-mc -triple x86_64-apple-macho -filetype obj -o - %s | llvm-readobj -sections | FileCheck %s
_label:
.cfi_startproc
.cfi_def_cfa_register rsp
.cfi_endproc
// CHECK: Section {
// CHECK: Name: __eh_frame
// CHECK: }