Re: Reentrant functions & concurrency

[Jeff Johnston <jjohnstn at redhat dot com>]

Schwarz, Konrad wrote:
> I have a question on the implementation of the functions in
> newlib/libc/reent, which are the reentrant versions of Unix system
> calls.
>
> Taking writer.c as an example, we find the following function:
>
> _ssize_t
> _DEFUN (_write_r, (ptr, fd, buf, cnt),
>      struct _reent *ptr _AND
>      int fd _AND 
>      _CONST _PTR buf _AND
>      size_t cnt)
> {
>   _ssize_t ret;
>
>   errno = 0;
>   if ((ret = (_ssize_t)_write (fd, buf, cnt)) == -1 && errno != 0)
>     ptr->_errno = errno;
>   return ret;
> } 
>
> Of particular interest are the three lines starting with "errno = 0;".
>
> What prevents an operating system from performing a context switch
> between the call to _write() and the following assignment of errno to
> the reentrancy structure?  Such a context switch could cause errno to
> change, which would result in an incorrect value of errno to be returned
> to the application.
>
> Shouldn't there be a _LOCK_T __errno_lock defined somewhere, a call to
> __lock_acquire (__errno_lock) just before the line "errno = 0;" and a
> call to __lock_release (__errno_lock) just before the line "return
> ret;"?
>
> Regards,
>
> Konrad Schwarz

This is a historical situation.  If you look at libc/include/reent.h you 
will see the following comment which acknowledges that supplying just 
underscore versions of syscalls won't be truly reentrant at the syscall 
level:

*********************************************************************
   2) Define namespace clean versions of the system calls by prefixing
      them with '_' (eg: _open, _close, etc.).  Technically, there won't
      be true reentrancy at the syscall level, but the library will be
      namespace clean.
*********************************************************************

Now that there are locks in newlib, adding them as mentioned above is a 
sound idea.  There is one modification required.  x86-linux specifies 
MISSING_SYSCALL_NAMES and the syscalls themselves are reentrant (i.e. 
they use normal errno, not the external one provided).  I am guessing 
that RTEMS would be in the same boat.  It would be useful to be able to 
turn the errno locking off via a flag (e.g. 
SYSCALLS_USE_REENTRANT_ERRNO) when it is not needed.  The locking macros 
would be indirectly specified via macros in a local.h header.  Would you 
like me to whip up something or do you already have a trial patch to modify?

-- Jeff J.