From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Message-ID: <559210FF.6030900@kernel.dk> Date: Mon, 29 Jun 2015 21:46:07 -0600 From: Jens Axboe MIME-Version: 1.0 Subject: Re: FreeBSD build broken following _Static_assert commit References: <5591C7F0.1090004@cran.org.uk> <55920381.6060906@kernel.dk> <55920FE0.60901@cran.org.uk> In-Reply-To: <55920FE0.60901@cran.org.uk> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit To: Bruce Cran , "fio@vger.kernel.org" List-ID: On 06/29/2015 09:41 PM, Bruce Cran wrote: > On 6/29/2015 8:48 PM, Jens Axboe wrote: >> Thanks for reporting this, Bruce. No idea why clang complains about >> that, it's definitely a constant expression. Anyway, committed this: >> >> http://git.kernel.dk/cgit/fio/commit/?id=94815a5c8366a9290167e8539f29994c2d43d15c >> >> >> which adds this sort of check to the configure _Static_assert() check. > > It seems the relevant text is: > > note: cast that performs the conversions of a reinterpret_cast is not > allowed in a constant expression But it's bogus, it's nothing but static pointer math. > Is there something about our use of offsetof that means we can't depend > on the version from ? No other reason than having a generic solution is superior to depending on whatever might be supplied with the given system. -- Jens Axboe