www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.bugs - [Issue 3956] New: linker removes underscore from all exported symbols of a module but the first

reply d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=3956

           Summary: linker removes underscore from all exported symbols of
                    a module but the first
           Product: D
           Version: unspecified
          Platform: Other
        OS/Version: Windows
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Optlink
        AssignedTo: nobody puremagic.com
        ReportedBy: r.sagitario gmx.de



PST ---
this test program:

module testexport;

export void foo1() {}
export void foo2() {}
export void foo3() {}

void main() {}

compiles, but exports in the resulting executable (not a DLL here for
simplicity) are wrong.

dumpbin /exports testexport.exe
[...] ordinal hint RVA name 1 0 00003014 D10testexport4foo2FZv 2 1 00003018 D10testexport4foo3FZv 3 2 00003010 _D10testexport4foo1FZv [...] The object file seems ok, here's a snippet from obj2asms' output: extrn _D10testexport4foo1FZv ;expdef expflag=x00, export '_D10testexport4foo1FZv', internal '', ordinal=x0 extrn _D10testexport4foo2FZv ;expdef expflag=x00, export '_D10testexport4foo2FZv', internal '', ordinal=x0 extrn _D10testexport4foo3FZv ;expdef expflag=x00, export '_D10testexport4foo3FZv', internal '', ordinal=x0 [...] public _D10testexport12__ModuleInfoZ _D10testexport4foo1FZv COMDAT flags=x0 attr=x0 align=x0 _D10testexport4foo2FZv COMDAT flags=x0 attr=x0 align=x0 _D10testexport4foo3FZv COMDAT flags=x0 attr=x0 align=x0 __Dmain COMDAT flags=x0 attr=x0 align=x0 -- Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email ------- You are receiving this mail because: -------
Mar 13 2010
next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=3956


Rainer Schuetze <r.sagitario gmx.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andrej.mitrovich gmail.com



PST ---
*** Issue 7186 has been marked as a duplicate of this issue. ***

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Dec 31 2011
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=3956


Martin Nowak <code dawg.eu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |dll
                 CC|                            |code dawg.eu
           Severity|normal                      |major



This just broke my shiny new loadFunc, loadSym implementation and it took me
+1h to get here. Any ideas how to fix this?

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Sep 27 2013
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=3956




Presumably something in this TWEAK_EXPORT_NAMES function goes wrong or the
function isn't called under certain circumstances. It seems simple enough to
remove the underscore striping altogether. Is it possible to debug optlink with
some debug infos. How to build this thing?

https://github.com/DigitalMars/optlink/blob/475bc5c1fa28eaf899ba4ac1dcfe2ab415db16c6/common/coment.asm#L529

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Sep 27 2013
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=3956




#ifdef fg_pe

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Sep 28 2013
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=3956




YES_EXPORT PROC
;; ...
#ifdef fg_pe
   if (OUTPUT_PE) // false on first test
      TWEAK_EXPORT_NAMES();
#endif
    DEFINE_EXPORT(); // could set ANY_USE32, but it's usually already
    SELECT_OUTPUT_SEGMENTED_OR_PE();
;; ...
YES_EXPORT ENDP

SELECT_OUTPUT_SEGMENTED_OR_PE PROC
    if (!(OUTPUT_SEGMENTED | OUTPUT_PE))
       return;
    if (ANY_USE32) // this is already set by defining the output segments
        SELECT_OUTPUT_PE(); // sets OUTPUT_PE
    else
        SELECT_OUTPUT_SEGMENTED();
SELECT_OUTPUT_SEGMENTED_OR_PE ENDP

I found the bug, thanks to OllyDbg, but I could use some help to interpret and
resolve it.
From what I understand optlink determines the output format OUTPUT_SEGMENTED or
OUTPUT_PE depending on the input files. That is when it first encounters some
usage of 32-bit symbols it will set ANY_USE32.
Striping the leading underscore in the exported name is only done for
OUTPUT_PE.
The problem comes from the circular dependency in the OUTPUT_PE test above. For
the first export symbol no output format has yet been set, thus the test fails
and the symbol is not stripped.

One way to resolve this would be to prematurely test for ANY_USE32 if no OUTPUT
is selected.

    if (OUTPUT_PE || !OUTPUT_SEGMENTED && ANY_USE32)


Any ideas?

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Sep 28 2013
prev sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=3956


Martin Nowak <code dawg.eu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |pull



https://github.com/DigitalMars/optlink/pull/10

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Sep 28 2013