Date: Wed, 17 Mar 2004 12:59:13 +0100 From: des@des.no (Dag-Erling =?iso-8859-1?q?Sm=F8rgrav?=) To: Ruslan Ermilov <ru@freebsd.org> Cc: current@freebsd.org Subject: Re: [current tinderbox] failure on ia64/ia64 Message-ID: <xzpish38z1q.fsf@dwp.des.no> In-Reply-To: <20040317113931.GF49920@ip.net.ua> (Ruslan Ermilov's message of "Wed, 17 Mar 2004 13:39:31 %2B0200") References: <20040317040254.386947303A@freebsd-current.sentex.ca> <xzpad2fhixn.fsf@dwp.des.no> <20040317113931.GF49920@ip.net.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
Ruslan Ermilov <ru@freebsd.org> writes:
> I cannot reproduce it either, but perhaps this is the same reason why
> we still have -Wno-uninitialized in <bsd.sys.mk> for normal WARNS
> levels.
No, we use -Wno-uninitialized because of cases where it is obvious to
humans but not to gcc that a variable is initialized, such as:
if (foo !=3D NULL)
bar =3D baz(foo);
/* some code that doesn't change foo */
if (foo !=3D NULL)
printf("%d\n", bar);
If you remove the code between the two ifs, gcc will coalesce them and
realize nothing fishy is going on (unless you compile with -O0).
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?xzpish38z1q.fsf>