From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.org/gmane.linux.lib.musl.general/7405 Path: news.gmane.org!not-for-mail From: Alexander Monakov Newsgroups: gmane.linux.lib.musl.general Subject: Re: Explicit casts in ctype.h suppress compiler warnings Date: Fri, 17 Apr 2015 21:24:09 +0300 (MSK) Message-ID: References: <1429289394.7038.3.camel@inria.fr> <20150417165238.GA6817@brightrain.aerifal.cx> Reply-To: musl@lists.openwall.com NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: ger.gmane.org 1429295066 27579 80.91.229.3 (17 Apr 2015 18:24:26 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Fri, 17 Apr 2015 18:24:26 +0000 (UTC) To: musl@lists.openwall.com Original-X-From: musl-return-7418-gllmg-musl=m.gmane.org@lists.openwall.com Fri Apr 17 20:24:26 2015 Return-path: Envelope-to: gllmg-musl@m.gmane.org Original-Received: from mother.openwall.net ([195.42.179.200]) by plane.gmane.org with smtp (Exim 4.69) (envelope-from ) id 1YjAwB-0005QY-EN for gllmg-musl@m.gmane.org; Fri, 17 Apr 2015 20:24:23 +0200 Original-Received: (qmail 21770 invoked by uid 550); 17 Apr 2015 18:24:21 -0000 Mailing-List: contact musl-help@lists.openwall.com; run by ezmlm Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: Original-Received: (qmail 21745 invoked from network); 17 Apr 2015 18:24:21 -0000 In-Reply-To: <20150417165238.GA6817@brightrain.aerifal.cx> User-Agent: Alpine 2.11 (LNX 23 2013-08-11) Xref: news.gmane.org gmane.linux.lib.musl.general:7405 Archived-At: > In another place (math.h) I removed this type of compound literal > usage because it was incompatible with C++, but the macros are > suppressed in C++ anyway. Still they might break -pedantic with > -std=c89. I do like this approach best in principle if it works > though, because the rules for when an error occurs are basically the > same as the rules for a real function. I confirm that the idea works, and as Rich said it causes a warning with -pedantic -std=c89 with gcc-4.5..4.7 (but not 4.8, 4.9). > Do you have an idea in mind for how we could achieve that? I suspect > the macros are still better optimizable than the inline function > approach, so I'd lean towards doing a macro that avoids evaluating c > and just checks its type, which would involve using ?: I think. I admit I was thinking of doing isspace-style inlines everywhere, but thanks to your suggestion I was able to come up with this: static __inline void __is_int(int a) {} #define isdigit(a) (__is_int(0?(a):0), ((unsigned)(a)-'0') < 10) Actually, thinking about GCC behavior a bit more, I was lucky that there's a warning for isspace in the first place: wrong argument to __isspace originates in the context of a system-header-declared macro, so generally I'd say that the same warning-suppression logic should have applied. Alexander