2011-06-16 17:14:38 +00:00
|
|
|
; RUN: llvm-as < %s | llvm-dis | FileCheck %s
|
2014-08-19 21:08:27 +00:00
|
|
|
; RUN: verify-uselistorder %s
|
2011-06-16 17:14:38 +00:00
|
|
|
|
|
|
|
|
!0 = metadata !{metadata !"zero"}
|
|
|
|
|
!1 = metadata !{metadata !"one"}
|
|
|
|
|
!2 = metadata !{metadata !"two"}
|
|
|
|
|
|
|
|
|
|
!foo = !{!0, !1, !2}
|
|
|
|
|
; CHECK: !foo = !{!0, !1, !2}
|
|
|
|
|
|
|
|
|
|
!\23pragma = !{!0, !1, !2}
|
|
|
|
|
; CHECK: !\23pragma = !{!0, !1, !2}
|
|
|
|
|
|
|
|
|
|
; \31 is the digit '1'. On emission, we escape the first character (to avoid
|
|
|
|
|
; conflicting with anonymous metadata), but not the subsequent ones.
|
|
|
|
|
!\31\31\31 = !{!0, !1, !2}
|
|
|
|
|
; CHECK: !\3111 = !{!0, !1, !2}
|
|
|
|
|
|
|
|
|
|
!\22name\22 = !{!0, !1, !2}
|
|
|
|
|
; CHECK: !\22name\22 = !{!0, !1, !2}
|
|
|
|
|
|
|
|
|
|
; \x doesn't mean anything, so we parse it literally but escape the \ into \5C
|
|
|
|
|
; when emitting it, followed by xfoo.
|
|
|
|
|
!\xfoo = !{!0, !1, !2}
|
|
|
|
|
; CHECK: !\5Cxfoo = !{!0, !1, !2}
|
