[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <alpine.LNX.2.11.1504211148400.2677@monopod.intra.ispras.ru>
Date: Tue, 21 Apr 2015 11:56:30 +0300 (MSK)
From: Alexander Monakov <amonakov@...ras.ru>
To: musl@...ts.openwall.com
Subject: Re: x86_64 and x32 fail to build
On Tue, 21 Apr 2015, Jens Gustedt wrote:
> Since such a solution would use a gcc'ish extension __typeof__,
> anyhow, perhaps it would be preferable to use another one, namely
> __builtin_choose_expr:
>
> #define __scc(X) __builtin_choose_expr(sizeof(1?(X):0ULL) < 8, (unsigned long) (X), (long long) (X))
>
> if that helps to quiencen the warning?
It doesn't help: the warning is still produced for the unevaluated branch of
__builtin_choose_expr, similarly to ternary operator (I tested with gcc-4.8).
My __scc1 could be written a bit simpler:
#define __scc1(X) (__typeof__((X)-(X))) (X)
but still it doesn't work for pointers to void or incomplete types.
Alexander
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
