[Bug target/32219] New: optimizer causes wrong code in pic/hidden/weak symbol checking.

[Bug target/32219] New: optimizer causes wrong code in pic/hidden/weak symbol checking.

[pluto at agmk dot net]

following testcase causes gpf on i386.

$ cat exe.c
extern void f() __attribute__((weak,visibility("hidden")));
extern int puts( char const* );
int main()
{
        f ? f() : puts( "f == null, skipped." );
        return 0;
}

$ gcc exe.c -fPIE --save-temps -m32 -O1 && ./a.out
Segmentation fault

this fails on i386 with gcc 3.4.5, 4.0, 4.1 and 4.2 (4.3 not tested).
x86_64 and ppc works fine, so it looks like a ix86 target bug.


-- 
           Summary: optimizer causes wrong code in pic/hidden/weak symbol
                    checking.
           Product: gcc
           Version: 4.2.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: pluto at agmk dot net
GCC target triplet: i386-gnu-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[pluto at agmk dot net]

------- Comment #1 from pluto at agmk dot net  2007-06-05 12:53 -------
btw, imho the weak+hidden is not a valid combination.
such symbol can't be resolved in runtime because it doesn't exist
in elf rel.plt/rel.dyn tables. it can be resolved only during
linking several objects into one piece (e.g. exec/shlib),
but in such case the weak+hidden behaves like plain hidden.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2007-06-05 13:30 -------
f always will bind local ...
Also you should be using -PIE when linking.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[pluto at agmk dot net]

------- Comment #3 from pluto at agmk dot net  2007-06-05 14:10 -------
(In reply to comment #2)

> Also you should be using -PIE when linking.

hmm, it doesn't work with int main();

$ gcc -s main.c -fpie -Wl,-pie
/usr/bin/ld: /usr/lib64/crt1.o: relocation R_X86_64_32S against
`__libc_csu_fini' can not be used when making a shared object;
 recompile with -fPIC
/usr/lib64/crt1.o: could not read symbols: Bad value
collect2: ld returned 1 exit status

huh, glibc bug, linker bug?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[pluto at agmk dot net]

------- Comment #4 from pluto at agmk dot net  2007-06-05 14:13 -------
(In reply to comment #2)
> f always will bind local ...

so, should gcc reject weak attribute in this (hidden visibility) case?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[pluto at agmk dot net]

------- Comment #5 from pluto at agmk dot net  2008-02-23 21:09 -------
(In reply to comment #3)
> (In reply to comment #2)
> 
> > Also you should be using -PIE when linking.
> 
> hmm, it doesn't work with int main();
> 
> $ gcc -s main.c -fpie -Wl,-pie
> /usr/bin/ld: /usr/lib64/crt1.o: relocation R_X86_64_32S against
> `__libc_csu_fini' can not be used when making a shared object;
>  recompile with -fPIC
> /usr/lib64/crt1.o: could not read symbols: Bad value
> collect2: ld returned 1 exit status
> 
> huh, glibc bug, linker bug?

with gcc-4.2.3 and binutils-2.18.50.0.4 it links fine.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[hjl dot tools at gmail dot com]

------- Comment #6 from hjl dot tools at gmail dot com  2008-02-23 23:12 -------
Weak and hidden aren't really compatible. Linker should enforce it.
I opened a linker bug:

http://sourceware.org/bugzilla/show_bug.cgi?id=5789

As for gcc, I think it is OK since it is a user error and linker
should issue an error. Besides, gcc doesn't know if the .o
file will be used for PIE/shared library or normal executable.
The runtime error only happens with PIE/shared library, not
normal executable.


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://sourceware.org/bugzil
                   |                            |la/show_bug.cgi?id=5789
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[hjl dot tools at gmail dot com]

------- Comment #7 from hjl dot tools at gmail dot com  2008-02-25 22:16 -------
It is a compiler bug after all. From:

http://groups.google.com/group/generic-abi/browse_thread/thread/4364eb484397ebe0

A hidden symbol must be defined in the same component, *if it is  
defined at all*. That last part is the key to the issue. In the case  
of a WEAK HIDDEN UNDEF symbol, it is possible for the symbol to remain  
undefined (or, looked at another way, it is possible for the symbol to  
be defined at link time as an absolute symbol with value 0). I agree  
with Daniel; the compiler should be aware of this possibility and  
generate code that will not require a dynamic relocation for this case.

For a weak and hidden symbol, it is bound local only if PIC is false or
it is used for branch. Otherwise, it should be treated with default
visibility.


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|INVALID                     |
            Version|4.2.1                       |4.3.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[castet dot matthieu at free dot fr]

------- Comment #8 from castet dot matthieu at free dot fr  2009-12-29 23:20 -------
What's the status of this bug ?
The same things can happen in libraries with fpic


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[aldot at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1171 bytes --]



------- Comment #9 from aldot at gcc dot gnu dot org  2010-03-08 19:28 -------
(In reply to comment #8)
> What's the status of this bug ?

we currently still end up with
call 0
on e.g. i386

> The same things can happen in libraries with fpic

yes. Thing is that we could theoretically work around it by explicitly looking
at the addr ¹) but that's just plain disgusting imho. And the hardened guys
will not like the idea to drop DOPIC (i.e. build members of .a without PIC).

So.. What's the status of that bug? Current binutils-2.20 and somewhat current
gcc basically generate "wrong" code, or at least code that behaves in an
unpleasant way causing grief..

¹) https://bugs.uClibc.org/1033

thanks and cheers,


-- 

aldot at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2010-03-08 19:28:29
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[rguenth at gcc dot gnu dot org]

------- Comment #10 from rguenth at gcc dot gnu dot org  2010-03-09 12:39 -------
Well, simply re-ordering the visibility and the weak check in
varasm.c:default_binds_local_p_1 should do the trick.


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rguenth at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[aldot at gcc dot gnu dot org]

------- Comment #11 from aldot at gcc dot gnu dot org  2010-03-16 13:35 -------
(In reply to comment #10)
> Well, simply re-ordering the visibility and the weak check in
> varasm.c:default_binds_local_p_1 should do the trick.
> 
It does.
http://gcc.gnu.org/ml/gcc-patches/2010-03/msg00665.html


-- 

aldot at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[ncopa at alpinelinux dot org]

------- Comment #12 from ncopa at alpinelinux dot org  2010-06-30 11:52 -------
whats the status on this bug? It still happens on gcc-4.4.4.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[aldot at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

--- Comment #13 from Bernhard Reutner-Fischer <aldot at gcc dot gnu.org> 2012-06-28 22:49:52 UTC ---
Created attachment 27716
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=27716
testcase

This was (accidentally) fixed on for at least 4.7 ff but without adding a
testcase like the attached.

[Bug target/32219] optimizer causes wrong code in pic/hidden/weak symbol checking.

[mikpe at it dot uu.se]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32219

Mikael Pettersson <mikpe at it dot uu.se> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mikpe at it dot uu.se

--- Comment #14 from Mikael Pettersson <mikpe at it dot uu.se> 2012-09-01 15:17:28 UTC ---
(In reply to comment #13)
> This was (accidentally) fixed on for at least 4.7 ff but without adding a
> testcase like the attached.

Are you sure it's been fixed? The test case segfaults for me on x86_64-linux
when compiled by 4.8-20120826, 4.7-20120825, or 4.6-20120810, with -O1 -m32 and
-fPIC or -fPIE. Without -m32 it doesn't segfault.