maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/DebugInfo/X86/multiple-at-const-val.ll

62 lines
4.2 KiB
LLVM
Raw Normal View History

[Dwarf] avoid emitting multiple AT_const_value for static memebers. Testing case is reduced from MultiSource/BenchMarks/Prolangs-C++/deriv1. rdar://problem/13071590 llvm-svn: 174235
2013-02-02 07:54:37 +08:00
; RUN: llc -O0 %s -mtriple=x86_64-apple-darwin -filetype=obj -o %t
; RUN: llvm-dwarfdump %t | FileCheck %s
; rdar://13071590
; Check we are not emitting mutliple AT_const_value for a single member.
; CHECK: .debug_info contents:
; CHECK: DW_TAG_compile_unit
; CHECK: DW_TAG_class_type
; CHECK: DW_TAG_member
; CHECK: badbit
; CHECK: DW_AT_const_value [DW_FORM_data4] (0x00000001)
; CHECK-NOT: DW_AT_const_value
; CHECK: NULL
%"class.std::basic_ostream" = type { i32 (...)**, %"class.std::basic_os" }
%"class.std::basic_os" = type { %"class.std::os_base", %"class.std::basic_ostream"*, i8, i8 }
%"class.std::os_base" = type { i32 (...)**, i64, i64, i32, i32, i32 }
@_ZSt4cout = external global %"class.std::basic_ostream"
@.str = private unnamed_addr constant [6 x i8] c"c is \00", align 1
Removing ssp and uwtable from the testcase llvm-svn: 174259
2013-02-02 09:34:38 +08:00
define i32 @main() {
[Dwarf] avoid emitting multiple AT_const_value for static memebers. Testing case is reduced from MultiSource/BenchMarks/Prolangs-C++/deriv1. rdar://problem/13071590 llvm-svn: 174235
2013-02-02 07:54:37 +08:00
entry:
%call1.i = tail call %"class.std::basic_ostream"* @test(%"class.std::basic_ostream"* @_ZSt4cout, i8* getelementptr inbounds ([6 x i8]* @.str, i64 0, i64 0), i64 5)
ret i32 0
}
declare %"class.std::basic_ostream"* @test(%"class.std::basic_ostream"*, i8*, i64)
declare void @llvm.dbg.value(metadata, i64, metadata) nounwind readnone
!llvm.dbg.cu = !{!0}
Revert "Revert "PR14606: debug info imported_module support"" This reverts commit r179840 with a fix to test/DebugInfo/two-cus-from-same-file.ll I'm not sure why that test only failed on ARM & MIPS and not X86 Linux, even though the debug info was clearly invalid on all of them, but this ought to fix it. llvm-svn: 179996
2013-04-22 14:12:31 +08:00
!0 = metadata !{i32 786449, i32 4, metadata !961, metadata !"clang version 3.3 (trunk 174207)", i1 true, metadata !"", i32 0, metadata !1, metadata !955, metadata !956, metadata !1786, metadata !1786, metadata !""} ; [ DW_TAG_compile_unit ] [/privite/tmp/student2.cpp] [DW_LANG_C_plus_plus]
Remove the (apparently) unnecessary debug info metadata indirection. The main lists of debug info metadata attached to the compile_unit had an extra layer of metadata nodes they went through for no apparent reason. This patch removes that (& still passes just as much of the GDB 7.5 test suite). If anyone can show evidence as to why these extra metadata nodes are there I'm open to reverting this patch & documenting why they're there. llvm-svn: 174266
2013-02-02 13:56:24 +08:00
!1 = metadata !{metadata !26}
Revert "Adding DIImportedModules to DIScopes." This reverts commit 342d92c7a0adeabc9ab00f3f0d88d739fe7da4c7. Turns out we're going with a different schema design to represent DW_TAG_imported_modules so we won't need this extra field. llvm-svn: 178215
2013-03-28 10:44:59 +08:00
!4 = metadata !{i32 786489, null, metadata !"std", metadata !5, i32 48} ; [ DW_TAG_namespace ]
!5 = metadata !{i32 786473, metadata !1801} ; [ DW_TAG_file_type ]
[Dwarf] avoid emitting multiple AT_const_value for static memebers. Testing case is reduced from MultiSource/BenchMarks/Prolangs-C++/deriv1. rdar://problem/13071590 llvm-svn: 174235
2013-02-02 07:54:37 +08:00
!25 = metadata !{i32 786472, metadata !"_S_os_fmtflags_end", i64 65536} ; [ DW_TAG_enumerator ]
Revert "Adding DIImportedModules to DIScopes." This reverts commit 342d92c7a0adeabc9ab00f3f0d88d739fe7da4c7. Turns out we're going with a different schema design to represent DW_TAG_imported_modules so we won't need this extra field. llvm-svn: 178215
2013-03-28 10:44:59 +08:00
!26 = metadata !{i32 786436, metadata !1801, metadata !4, metadata !"_Ios_Iostate", i32 146, i64 32, i64 32, i32 0, i32 0, null, metadata !27, i32 0, i32 0} ; [ DW_TAG_enumeration_type ]
[Dwarf] avoid emitting multiple AT_const_value for static memebers. Testing case is reduced from MultiSource/BenchMarks/Prolangs-C++/deriv1. rdar://problem/13071590 llvm-svn: 174235
2013-02-02 07:54:37 +08:00
!27 = metadata !{metadata !28, metadata !29, metadata !30, metadata !31, metadata !32}
!28 = metadata !{i32 786472, metadata !"_S_goodbit", i64 0} ; [ DW_TAG_enumerator ] [_S_goodbit :: 0]
!29 = metadata !{i32 786472, metadata !"_S_badbit", i64 1} ; [ DW_TAG_enumerator ] [_S_badbit :: 1]
!30 = metadata !{i32 786472, metadata !"_S_eofbit", i64 2} ; [ DW_TAG_enumerator ] [_S_eofbit :: 2]
!31 = metadata !{i32 786472, metadata !"_S_failbit", i64 4} ; [ DW_TAG_enumerator ] [_S_failbit :: 4]
!32 = metadata !{i32 786472, metadata !"_S_os_ostate_end", i64 65536} ; [ DW_TAG_enumerator ] [_S_os_ostate_end :: 65536]
Revert "Adding DIImportedModules to DIScopes." This reverts commit 342d92c7a0adeabc9ab00f3f0d88d739fe7da4c7. Turns out we're going with a different schema design to represent DW_TAG_imported_modules so we won't need this extra field. llvm-svn: 178215
2013-03-28 10:44:59 +08:00
!49 = metadata !{i32 786434, metadata !1801, metadata !4, metadata !"os_base", i32 200, i64 1728, i64 64, i32 0, i32 0, null, metadata !50, i32 0, metadata !49, null} ; [ DW_TAG_class_type ]
[Dwarf] avoid emitting multiple AT_const_value for static memebers. Testing case is reduced from MultiSource/BenchMarks/Prolangs-C++/deriv1. rdar://problem/13071590 llvm-svn: 174235
2013-02-02 07:54:37 +08:00
!50 = metadata !{metadata !77}
Revert "Adding DIImportedModules to DIScopes." This reverts commit 342d92c7a0adeabc9ab00f3f0d88d739fe7da4c7. Turns out we're going with a different schema design to represent DW_TAG_imported_modules so we won't need this extra field. llvm-svn: 178215
2013-03-28 10:44:59 +08:00
!54 = metadata !{i32 786453, i32 0, metadata !"", i32 0, i32 0, i64 0, i64 0, i64 0, i32 0, null, metadata !55, i32 0, i32 0} ; [ DW_TAG_subroutine_type ]
[Dwarf] avoid emitting multiple AT_const_value for static memebers. Testing case is reduced from MultiSource/BenchMarks/Prolangs-C++/deriv1. rdar://problem/13071590 llvm-svn: 174235
2013-02-02 07:54:37 +08:00
!55 = metadata !{metadata !56}
Revert "Adding DIImportedModules to DIScopes." This reverts commit 342d92c7a0adeabc9ab00f3f0d88d739fe7da4c7. Turns out we're going with a different schema design to represent DW_TAG_imported_modules so we won't need this extra field. llvm-svn: 178215
2013-03-28 10:44:59 +08:00
!56 = metadata !{i32 786468, null, null, metadata !"int", i32 0, i64 32, i64 32, i64 0, i32 0, i32 5} ; [ DW_TAG_base_type ]
!77 = metadata !{i32 786445, metadata !1801, metadata !49, metadata !"badbit", i32 331, i64 0, i64 0, i64 0, i32 4096, metadata !78, i32 1} ; [ DW_TAG_member ]
!78 = metadata !{i32 786470, null, null, metadata !"", i32 0, i64 0, i64 0, i64 0, i32 0, metadata !79} ; [ DW_TAG_const_type ]
!79 = metadata !{i32 786454, metadata !49, metadata !"ostate", metadata !5, i32 327, i64 0, i64 0, i64 0, i32 0, metadata !26} ; [ DW_TAG_typedef ]
Remove the (apparently) unnecessary debug info metadata indirection. The main lists of debug info metadata attached to the compile_unit had an extra layer of metadata nodes they went through for no apparent reason. This patch removes that (& still passes just as much of the GDB 7.5 test suite). If anyone can show evidence as to why these extra metadata nodes are there I'm open to reverting this patch & documenting why they're there. llvm-svn: 174266
2013-02-02 13:56:24 +08:00
!955 = metadata !{i32 0}
!956 = metadata !{metadata !960}
Revert "Adding DIImportedModules to DIScopes." This reverts commit 342d92c7a0adeabc9ab00f3f0d88d739fe7da4c7. Turns out we're going with a different schema design to represent DW_TAG_imported_modules so we won't need this extra field. llvm-svn: 178215
2013-03-28 10:44:59 +08:00
!960 = metadata !{i32 786478, i32 0, metadata !961, metadata !"main", metadata !"main", metadata !"", metadata !961, i32 73, metadata !54, i1 false, i1 true, i32 0, i32 0, null, i32 256, i1 true, i32 ()* @main, null, null, metadata !955, i32 73} ; [ DW_TAG_subprogram ]
!961 = metadata !{i32 786473, metadata !1802} ; [ DW_TAG_file_type ]
Remove the (apparently) unnecessary debug info metadata indirection. The main lists of debug info metadata attached to the compile_unit had an extra layer of metadata nodes they went through for no apparent reason. This patch removes that (& still passes just as much of the GDB 7.5 test suite). If anyone can show evidence as to why these extra metadata nodes are there I'm open to reverting this patch & documenting why they're there. llvm-svn: 174266
2013-02-02 13:56:24 +08:00
!1786 = metadata !{metadata !1800}
[Dwarf] avoid emitting multiple AT_const_value for static memebers. Testing case is reduced from MultiSource/BenchMarks/Prolangs-C++/deriv1. rdar://problem/13071590 llvm-svn: 174235
2013-02-02 07:54:37 +08:00
!1800 = metadata !{i32 786484, i32 0, metadata !5, metadata !"badbit", metadata !"badbit", metadata !"badbit", metadata !5, i32 331, metadata !78, i32 1, i32 1, i32 1, metadata !77} ; [ DW_TAG_variable ]
Split out filename & directory from DIFile to start generalizing over DIScopes This is the first step to making all DIScopes have a common metadata prefix (so that things (using directives, for example) that can appear in any scope can be added to that common prefix). DIFile is itself a DIScope so the common prefix of all DIScopes cannot be a DIFile - instead it's the raw filename/directory name pair. llvm-svn: 177239
2013-03-18 05:13:55 +08:00
!1801 = metadata !{metadata !"os_base.h", metadata !"/privite/tmp"}
!1802 = metadata !{metadata !"student2.cpp", metadata !"/privite/tmp"}