From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============1621772032391002066==" MIME-Version: 1.0 From: Mark Wielaard To: elfutils-devel@lists.fedorahosted.org Subject: Re: [PATCH 1/2] Add is_executable to Dwfl_Module. Date: Wed, 10 Sep 2014 20:56:48 +0200 Message-ID: <1410375408.27502.25.camel@bordewijk.wildebeest.org> In-Reply-To: 20140910072127.GA1967@host2.jankratochvil.net --===============1621772032391002066== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Wed, 2014-09-10 at 09:21 +0200, Jan Kratochvil wrote: > On Tue, 09 Sep 2014 23:55:34 +0200, Roland McGrath wrote: > > > int segment; /* Index of first segment table entry. */ > > > bool gc; /* Mark/sweep flag. */ > > > + bool is_executable : 1; /* Use Dwfl::executable_for_core? */ > > = > > If you're going to use a bitfield, then make every other bool in that > > struct a bitfield too. But it's not usually worthwhile. It's > > premature microoptimization that privileges memory over CPU, which > > might not even be the right tradeoff any more. > = > I really do not mind, up to the maintainer how it all should be. I don't think it is very helpful or productive to refuse to have a technical opinion on a fair question about a code change you are proposing. But if you need a "ruling" to move forward then I am with Roland that the usage of booleans in a structure should be consistent. Don't introduce a one-off bool bitfield unless you group the other bools in the struct together and make them all bitfields. If you are not doing that, then just make it a regular bool field. Thanks, Mark --===============1621772032391002066==--