[Bug tools/21247] New: addr2line doesn't properly resolve filename/lineno

[Bug tools/21247] New: addr2line doesn't properly resolve filename/lineno

[marat at slonopotamus dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=21247

            Bug ID: 21247
           Summary: addr2line doesn't properly resolve filename/lineno
           Product: elfutils
           Version: unspecified
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: tools
          Assignee: unassigned at sourceware dot org
          Reporter: marat at slonopotamus dot org
                CC: elfutils-devel at sourceware dot org
  Target Milestone: ---

Way to reproduce:

1. wget http://slonopotamus.org/libUE4Editor-Core.so (warning, 50MB)
2. ./src/addr2line -f -e libUE4Editor-Core.so 0x2c09b2

Expected output (as produced by addr2line from GNU binutils and I know that
this is correct):

_Z20PlatformCrashHandleriP9siginfo_tPv
/home/marat/production/ue4/Engine/Source/Runtime/Core/Private/Linux/LinuxPlatformCrashContext.cpp:595

Actual output (tested against commit e22cc80923f94e54c9384421904461958899db5b):

_Z20PlatformCrashHandleriP9siginfo_tPv
??:0

So, function name is resolved correctly but filename/lineno info is missing.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug tools/21247] addr2line doesn't properly resolve filename/lineno

[marat at slonopotamus dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=21247

--- Comment #1 from Marat Radchenko <marat at slonopotamus dot org> ---
Additional info on various addr2line implementations misbehaving on this
testcase: https://github.com/gimli-rs/addr2line/issues/35#issue-213881844

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug tools/21247] addr2line doesn't properly resolve filename/lineno

[marat at slonopotamus dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=21247

--- Comment #2 from Marat Radchenko <marat at slonopotamus dot org> ---
gimli was fixed ( https://github.com/gimli-rs/addr2line/pull/36 ) and now is
consistent with binutils (and llvm-symbolizer). I also sent a patch to
elftoolchain so it also resolves filename/lineno properly (though not function
name yet): https://sourceforge.net/p/elftoolchain/tickets/545/#6d97

gimpi problem was that they didn't handle DW_AT_MIPS_linkage_name and
DW_AT_specification.

elftoolchain problem was that they din't handle DW_AT_ranges for compile units
(and only looked at lo/hi addresses). I suspect that you do not resolve
filename/lineno properly for the same reason.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug tools/21247] addr2line doesn't properly resolve filename/lineno

[mark at klomp dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=21247

Mark Wielaard <mark at klomp dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mark at klomp dot org

--- Comment #3 from Mark Wielaard <mark at klomp dot org> ---
Dumping the whole decoded line table with eu-readelf --debug-dump=decodedline
shows:

 
/home/marat/production/ue4/Engine/Source/Runtime/Core/Private/Linux/LinuxPlatformCrashContext.cpp
(mtime: 0, length: 0)
   593:6   S        0   0  0 +0x00000000002c099f
<_Z20PlatformCrashHandleriP9siginfo_tPv+0xef>
   593:6            0   0  0 +0x00000000002c09ae
<_Z20PlatformCrashHandleriP9siginfo_tPv+0xfe>
   595:3   S        0   0  0 +0x00000000002c09b0
<_Z20PlatformCrashHandleriP9siginfo_tPv+0x100>
   600:3   S        0   0  0 +0x00000000002c09b4
<_Z20PlatformCrashHandleriP9siginfo_tPv+0x104>
    75:1   S        0   0  0 +0x00000000002c09b9
<_Z20PlatformCrashHandleriP9siginfo_tPv+0x109>
    76:6   S        0   0  0 +0x00000000002c09be
<_Z20PlatformCrashHandleriP9siginfo_tPv+0x10e>

So there we are able to also deduce that address 0x2c09b2 is associated with
LinuxPlatformCrashContext.cpp:595:3

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug tools/21247] addr2line doesn't properly resolve filename/lineno

[mark at klomp dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=21247

--- Comment #4 from Mark Wielaard <mark at klomp dot org> ---
Looking at the debug_info we do have:

 [39ca09]    subprogram
             low_pc               (addr) +0x00000000002c08b0
<_Z20PlatformCrashHandleriP9siginfo_tPv>
             high_pc              (addr) +0x00000000002c0a05
             lo_user+0x1fe7       (flag) Yes
             frame_base           (block1) 
              [   0] reg7
             MIPS_linkage_name    (strp)
"_Z20PlatformCrashHandleriP9siginfo_tPv"
             name                 (strp) "PlatformCrashHandler"
             decl_file            (data1) 107
             decl_line            (data2) 580
             external             (flag) Yes
             lo_user+0x1fe1       (flag) Yes

Which is inside a compile_unit generated by clang:

 [29d634]  compile_unit
           producer             (strp) "clang version 3.7.1
(tags/RELEASE_371/fi
nal)"
           language             (data2) C_plus_plus (4)
           name                 (strp) "-"
           stmt_list            (data4) 581678
           comp_dir             (strp)
"/home/marat/production/ue4/Engine/Source
"
           lo_user+0x1fe1       (flag) Yes
           low_pc               (addr) +000000000000000000
           ranges               (data4) range list [ c92e0]

Note that the low_pc seems bogus (zero). And the range list seems bogus too (it
consists of 800+ entries some of which look bogus too
(+0x0000000000000001..+0x0000000000000001) and none seem to actually cover that
function addresses.

See also this discussion on the mailinglist:
https://sourceware.org/ml/elfutils-devel/2017-q2/msg00190.html

-- 
You are receiving this mail because:
You are on the CC list for the bug.