Bug: Missing va_end() in cygwin_internal()

Thu Dec 30 21:08:00 GMT 2004

Dave Korn wrote:
>> -----Original Message-----
>> From: cygwin-owner On Behalf Of Max Bowsher
>> Sent: 30 December 2004 19:05
>>> Christopher Faylor wrote:
>>> On Thu, Dec 30, 2004 at 05:59:02PM -0000, Max Bowsher wrote:
>>>> There is no va_end() call in cygwin_internal().
>>>>>> Or in functions from:
>>>>>> bsdlib.cc
>>> fork.cc
>>> pinfo.cc
>>> strace.cc
>>>>>> The lack doesn't seem to be causing much of a problem...
>>>> Standards say va_end is required, but may be a noop in some
>> implementations.
>> 99.9% of all known implementations. I guess it might be an issue if one 
> day
> we want to port cygwin to run on Vax or somesuch....
>>> I peered at the guts of gcc trying to work out whether that
>> was the case for cygwin, but got too lost to be sure.
>> AFAICS gcc supplies the mechanics to implement va_start and va_arg 
> however
> your architecture requires, but no way to define any functionality for 
> va_end
> at all.
>> Ah, found it. EXPAND_BUILTIN_VA_END was removed by
>> http://gcc.gnu.org/ml/gcc-patches/2004-01/msg01368.html
>> I guess that VAX port will be ok after all, since it's being a supported
> architecture for gcc means that it must not need va_end....

I guess there's little point being picky in following the standard, given 
that evidence.
Thanks for the information!
Max.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/