On 25 April 2012 23:15, Andreas Färber <andreas.faer...@web.de> wrote: > From: Andreas Färber <afaer...@suse.de> > > Based on the following Coccinelle patch: > > @@ > typedef int16, int_fast16_t; > @@ > -int16 > +int_fast16_t > > Avoids a workaround for AIX. > > Add typedef for pre-10 Solaris. > > Signed-off-by: Andreas Färber <afaer...@suse.de> > Cc: malc <av1...@comtv.ru> > Cc: Ben Taylor <bentaylor.sol...@gmail.com>
Reviewed-by: Peter Maydell <peter.mayd...@linaro.org> This is a pretty safe change, because the only external-to-softfloat APIs this changes are a couple used only by the ARM target (which I've tested and am happy are OK). The ARM vfp/neon tests I've run should have exercised the rest of the softfloat code reasonably well so the internal-to-softfloat changes should be OK too. In any case the type wasn't exactly-16-bits before so we already know the code is OK with the type being potentially wider than 16 bits. -- PMM
