https://bugs.kde.org/show_bug.cgi?id=433497

Mark Wielaard <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[email protected]

--- Comment #1 from Mark Wielaard <[email protected]> ---
wait and waitpid are implemented on top of the wait4 system call. Handling in
valgrind is defined in coregrind/m_syswrap/syswrap-generic.c. The current POST
handler is:

POST(sys_wait4)
{
   if (ARG2 != (Addr)NULL)
      POST_MEM_WRITE( ARG2, sizeof(int) );
   if (ARG4 != (Addr)NULL)
      POST_MEM_WRITE( ARG4, sizeof(struct vki_rusage) );
}

This would be the place to implement such logic.

I think checking if safetoderef *ARG2 & 0xff and if zero only define the lower
bits might work. But I am not 100% sure all wait status (exited, signaled,
stopped, continued) can be determined from the lower bits.

-- 
You are receiving this mail because:
You are watching all bug changes.

Reply via email to