X-Authentication-Warning: delorie.com: mail set sender to djgpp-workers-bounces using -f X-Authentication-Warning: envy.delorie.com: dj set sender to dj AT delorie DOT com using -f From: DJ Delorie To: djgpp-workers AT delorie DOT com Subject: Re: Missing functions in DJGPP libc (and/or) libm In-Reply-To: (djgpp-workers@delorie.com) Date: Mon, 13 Apr 2020 01:10:28 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain Reply-To: djgpp-workers AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk Andris Pavenis (andris DOT pavenis AT iki DOT fi) > https://gcc.gnu.org/git?p=gcc.git;a=commit;h=57391ddaf39f7cb85825c32e83feb1435889da51 > > less than a week ago. Should such change go in so close to release? Maybe no one can yet noticed > that it could break build for all targets for which Newlib is being used. Well, it's the "we depend on a c99 math library" choice that's the key here. If *that* decision was made long ago, and it just didn't get noticed until now, that's different than deciding a week ago to require c99. Either way it should be brought up on the gcc mailing list (or gcc's fortran mailing list?) to decide if the bug is "gcc assumes too much" or "targets need to upgrade". I don't follow gcc development as closely as I used to, now that I'm not doing gcc ports every day...