Date: Tue, 02 Dec 2008 13:55:10 +0100 From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= <des@des.no> To: Peter Wemm <peter@FreeBSD.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r185553 - head/lib/libutil Message-ID: <86r64q4tzl.fsf@ds4.des.no> In-Reply-To: <200812021010.mB2AAowJ055809@svn.freebsd.org> (Peter Wemm's message of "Tue, 2 Dec 2008 10:10:50 %2B0000 (UTC)") References: <200812021010.mB2AAowJ055809@svn.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Peter Wemm <peter@FreeBSD.org> writes: > Log: > Attempt a quick bandaid for arm build breakage. I went to the trouble = of > maintaining alignment, but I'm not sure how to tell gcc this. There's no good way to do this. In an ideal world, you could add an assertion that the offset is a multiple of __alignof__(struct kinfo_foo), and gcc would understand, but I don't think that works in practice. Coverity will probably complain about this as well, but hopefully the cast will pacify it like it pacifies gcc. In pw_util and gr_util, scf@ and I solved a similar issue by using a wrapper struct, but that wouldn't work here. DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86r64q4tzl.fsf>