maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/CodeGenCXX/globalinit-loc.cpp

24 lines
607 B
C++
Raw Normal View History

Debug info: (Bugfix) Make sure artificial functions like _GLOBAL__I_a are not associated with any source lines. Previously, if the Location of a Decl was empty, EmitFunctionStart would just keep using CurLoc, which would sometimes be correct (e.g., thunks) but in other cases would just point to a hilariously random location. This patch fixes this by completely eliminating all uses of CurLoc from EmitFunctionStart and rather have clients explicitly pass in a SourceLocation for the function header and the function body. rdar://problem/14985269 llvm-svn: 205999
2014-04-11 07:21:53 +08:00
// RUN: %clang_cc1 -emit-llvm -g %s -o - | FileCheck %s
// rdar://problem/14985269.
//
// Verify that the global init helper function does not get associated
// with any source location.
//
Include translation unit filename in global ctor symbol names. This makes it easier to see where a global ctor comes from, and it also makes ASan's init order analyzer output easier to understand. gcc does this too, but only in -fPIC mode for some reason. Don't do this for constructors with explicit init priority. Also prepend "sub_" before the 'I', that way regular constructors stay lexicographically after symbols with init priority (because ord('s') > ord('I')). gold seems to ignore the name of constructor symbols, and ld only looks at the symbol if it includes an init priority, which this patch doesn't change. Before: __GLOBAL_I_a Now: __GLOBAL_sub_I_myfile.cc llvm-svn: 208128
2014-05-07 04:32:45 +08:00
// CHECK: define internal void @_GLOBAL__sub_I_globalinit_loc.cpp
Follow-up to r205999: Emit an artificial location (valid scope, line 0) for CXXGlobalInit/Dtor helper functions. This makes _GLOBAL__I_a regain its DW_AT_high/low_pc in the debug info. Thanks to echristo for catching this! llvm-svn: 206088
2014-04-12 07:45:01 +08:00
// CHECK: !dbg ![[DBG:.*]]
Revert "Revert "DI: LLVM schema change: fold constants into string"" This reverts commit r218917, effectively reapplying r218913. Original commit message follows. -- Update debug info testcases for an LLVM metadata schema change to fold metadata constant operands into a single `MDString`. Part of PR17891. llvm-svn: 219011
2014-10-04 04:01:52 +08:00
// CHECK: metadata !{metadata !"0x2e\00\00\00_GLOBAL__sub_I_globalinit_loc.cpp\000\00{{.*}}\000", {{.*}} ; [ DW_TAG_subprogram ] [line 0] [local] [def]
Follow-up to r205999: Emit an artificial location (valid scope, line 0) for CXXGlobalInit/Dtor helper functions. This makes _GLOBAL__I_a regain its DW_AT_high/low_pc in the debug info. Thanks to echristo for catching this! llvm-svn: 206088
2014-04-12 07:45:01 +08:00
// CHECK: ![[DBG]] = metadata !{i32 0, i32 0,
Debug info: (Bugfix) Make sure artificial functions like _GLOBAL__I_a are not associated with any source lines. Previously, if the Location of a Decl was empty, EmitFunctionStart would just keep using CurLoc, which would sometimes be correct (e.g., thunks) but in other cases would just point to a hilariously random location. This patch fixes this by completely eliminating all uses of CurLoc from EmitFunctionStart and rather have clients explicitly pass in a SourceLocation for the function header and the function body. rdar://problem/14985269 llvm-svn: 205999
2014-04-11 07:21:53 +08:00
# 99 "someheader.h"
class A {
public:
A();
int foo() { return 0; }
};
# 5 "main.cpp"
A a;
int f() {
return a.foo();
}