[X86] Always generate precise CFA adjustments.
authorMichael Kuperstein <michael.m.kuperstein@intel.com>
Sun, 6 Dec 2015 13:06:20 +0000 (13:06 +0000)
committerMichael Kuperstein <michael.m.kuperstein@intel.com>
Sun, 6 Dec 2015 13:06:20 +0000 (13:06 +0000)
commitf7fc15ed79ddcb8cb26f00b61d32b757c62a24b3
tree708100ac03f3a70fe3c0ebddd55d60ba448a65a6
parente407ee0520400e5fa7316aed15caf788990de2af
[X86] Always generate precise CFA adjustments.

This removes the code path that generate "synchronous" (only correct at call site) CFA.
We will probably want to re-introduce it once we are capable of emitting different
.eh_frame and .debug_frame sections.

Differential Revision: http://reviews.llvm.org/D14948

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@254874 91177308-0d34-0410-b5e6-96231b3b80d8
include/llvm/CodeGen/MachineModuleInfo.h
lib/Target/X86/X86CallFrameOptimization.cpp
lib/Target/X86/X86FrameLowering.cpp
lib/Target/X86/X86MCInstLower.cpp
test/CodeGen/X86/push-cfi.ll
test/CodeGen/X86/tls-pie.ll
test/CodeGen/X86/win32-pic-jumptable.ll