Re: gcc auto-omit-frame-pointer vs msvc longjmp

[Kai Tietz <ktietz70@googlemail.com>]

2011/10/20 xunxun <xunxun1982@gmail.com>:
> Hi, all
>
> I think this issue causes the gdb crash on XP.
> You can see the thread: http://sourceware.org/ml/gdb/2011-10/msg00056.html
>
> My many friends and I can reproduce this crash issue, but no problem on Win7.
>
> On Thu, Oct 20, 2011 at 5:05 AM, Bob Breuer <breuerr@mc.net> wrote:
>> Kai Tietz wrote:
>>> 2011/10/18 Bob Breuer <breuerr@mc.net>:
>>>> Kai Tietz wrote:
>>>>> 2011/10/17 Bob Breuer <breuerr@mc.net>:
>>>>>> Richard Henderson wrote:
>>>>>>> On 10/17/2011 07:09 AM, Bob Breuer wrote:
>>>>>>>> Google finds a mention of longjmp failing with -fomit-frame-pointer:
>>>>>>>> http://lua-users.org/lists/lua-l/2005-02/msg00158.html
>>>>>>>>
>>>>>>>> Looks like gcc 4.6 turns on -fomit-frame-pointer by default.
>>>>>>> Hmm.  This is the first I've heard of a longjmp implementation
>>>>>>> failing without a frame pointer.  Presumably this is with the
>>>>>>> mingw i.e. msvc libc?
>>>>>> Yeah, mingw from www.mingw.org which I believe uses msvcrt.dll, package
>>>>>> gcc-core-4.6.1-2-mingw32-bin.
>>>>>>
>>>>>>> This is something that could be worked around in gcc, I suppose.
>>>>>>> We recognize longjmp for some things, we could force the use of
>>>>>>> a frame pointer for msvc targets too.
>>>>>>>
>>>>>>> For now it might be best to simply force -fno-omit-frame-pointer
>>>>>>> for mingw host in the configure script.
>>>>>> Here's a testcase that crashes on the longjmp:
>>>>>>
>>>>>> #include <stdio.h>
>>>>>> #include <setjmp.h>
>>>>>>
>>>>>> jmp_buf env;
>>>>>>
>>>>>> int test(void)
>>>>>> {
>>>>>>  int i;
>>>>>>
>>>>>>  asm("xor %%ebp,%%ebp" ::: "ebp");
>>>>>>
>>>>>>  i = setjmp(env);
>>>>>>  printf("i = %d\n", i);
>>>>>>
>>>>>>  if (i == 0)
>>>>>>    longjmp(env, 2);
>>>>>>
>>>>>>  return i;
>>>>>> }
>>>>>>
>>>>>> int main(void)
>>>>>> {
>>>>>>  return test();
>>>>>> }
>>>>>>
>>>>>> Remove the asm statement to make it not crash.  Obviously with
>>>>>> omit-frame-pointer, gcc can shove anything into ebp.
>>>>>>
>>>>>> Bob
>>>>> This crash isn'r related to ebp existing, or not. The issue is the
>>>>> hidden argument of setjmp, which is missing.  If you can try the
>>>>> following at top of file after include section.
>>>>>
>>>>> #define setjmp(BUF) _setjmpex((BUF), NULL)
>>>>> int __cdecl __attribute__ ((__nothrow__,__returns_twice__))
>>>>> _setjmp3(jmp_buf _Buf, void *_Ctx);
>>>>> ...
>>>> Did you mean _setjmp3 instead of _setjmpex?  With _setjmp3, it works
>>>> without the asm, but still crashes if I zero out ebp before the setjmp.
>>>>  Aren't the function arguments on the stack anyway?
>>>
>>> Yes, I mean _setjmp3 (pasto from headers and missed the second line
>>> prototyping _setjmp3).
>>> I repeat myself here.  setjmp() has an hidden arguement, which is
>>> passed on x86 on stack.  By not passing this required argument, setjmp
>>> will take a random-value from stack.  In your case 'i'.  btw if you
>>> would pre-initialize 'i' with zero, I would assume you won't see a
>>> crash, but anyway this is just by chance.
>>> For this I suggest to use here _setjmp3 instead, as here
>>> second-argument is documented as being present.
>>>
>>> Btw I tested your code with i686-pc-mingw32 version 4.6.x and 4.7.x
>>> gcc version.  With my suggested pattern, I don't see a crash for your
>>> provide test-code with, or without zero-ing ebp.
>>
>>
>> We probably have a difference in build or run environment.  I've
>> double-checked with another machine and can get the same crash in
>> longjmp when running the test executable on both WinXP and Win2k, but
>> not on Win7.  So it looks like Microsoft may have changed this "feature"
>> somewhere between WinXP and Win7.
>>
>> The msvcrt implementation of longjmp (or at least the one I'm looking
>> at) does a ebp based access using the saved value of ebp.  Here's the
>> relevant disassembly of longjmp:
>>
>> 0x7801e6f3 in longjmpex () from C:\WINNT\system32\msvcrt.dll
>> (gdb) disas
>> Dump of assembler code for function longjmpex:
>>   0x7801e6ef <+0>:     mov    0x4(%esp),%ebx
>> => 0x7801e6f3 <+4>:     mov    (%ebx),%ebp
>> ...
>>   0x7801e73d <+78>:    call   0x7800bd5e <abnormal_termination+56>
>> ...
>>   0x7800bd5e <+56>:    push   %ebx
>>   0x7800bd5f <+57>:    push   %ecx
>>   0x7800bd60 <+58>:    mov    $0x7803dc64,%ebx
>> => 0x7800bd65 <+63>:    mov    0x8(%ebp),%ecx
>>
>> It crashes on the access of 0x8(%ebp).  Those are the only 2 places
>> where this version of longjmp touches ebp.  Is it possible to force a
>> stackframe by just adding a suitable attribute to either the setjmp
>> function prototype, or the function which calls setjmp?
>>
>> Bob
>>
>
>
>
> --
> Best Regards,
> xunxun

This now makes sense. I use here Vista 64-bit, and Win7 64-bit and I
didn't found the issue.  But well, it is indeed related to different
msvcrt-version.

So there might be some need to have for a function using setjmp the
frame-pointer enabled.  I can confirm this by an older msvcrt.dll
version on my 64-bit box, too.

So bug can be re-opened.

Thanks,
Kai