From: Erik Faye-Lund Date: Thu, 24 Feb 2011 15:57:04 GMT Subject: Re: [RFC/PATCH 1/8] compat: provide a fallback va_copy definition Message-ID: In-Reply-To: <20110224142647.GA15477@sigill.intra.peff.net> On Thu, Feb 24, 2011 at 3:26 PM, Jeff King wrote: > va_copy is C99. Prior to this, the usual procedure was to > simply copy the va_list by assignment. > > Signed-off-by: Jeff King > --- > We have avoided using va_copy many times in the past, which has led to a > bunch of cut-and-paste. From everything I found searching the web, > implementations have historically either provided va_copy or just let > your code assume that simple assignment of worked. I couldn't find any > mention of any other alternatives. > > So my guess is that this will be sufficient, but I we won't really know > for sure until somebody reports a problem. :( > >  git-compat-util.h |    4 ++++ >  1 files changed, 4 insertions(+), 0 deletions(-) > > diff --git a/git-compat-util.h b/git-compat-util.h > index 9c23622..00d41e4 100644 > --- a/git-compat-util.h > +++ b/git-compat-util.h > @@ -535,6 +535,10 @@ void git_qsort(void *base, size_t nmemb, size_t size, >  #define fstat_is_reliable() 1 >  #endif > > +#ifndef va_copy > +#define va_copy(dst,src) (dst) = (src) > +#endif > + Wouldn't it be even more portable to fall back on use __va_copy (if present), as suggested by Junio in <7vbpip86q5.fsf@alter.siamese.dyndns.org>? He also suggested using memcpy instead of assignment in the same e-mail, due to a recommendation in the Autoconf manual. There's already a va_copy fall-back in compat/msvc.h, perhaps this should be removed?