Date: Tue, 9 Jul 2013 16:14:52 -0700 From: Jordan Hubbard <jordan.hubbard@gmail.com> To: John Baldwin <jhb@freebsd.org> Cc: "freebsd-arch@freebsd.org" <freebsd-arch@freebsd.org> Subject: Re: General purpose library for name/value pairs. Message-ID: <CC5D1844-EA9A-4B88-A842-F7F4307458ED@gmail.com> In-Reply-To: <201307091039.26835.jhb@freebsd.org> References: <20130704215329.GG1402@garage.freebsd.pl> <20130708213351.GB1405@garage.freebsd.pl> <D2E98A8F-F765-4A56-96CD-4410944A2910@turbofuzz.com> <201307091039.26835.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Yes, for anonymous shms this is certainly true. Hey, if the application pro= grammer *wants* to just pass a fd reference then there's certainly nothing p= reventing them from doing so in either case, I'm just suggesting that having= an explicit OOB type for shared memory (named or otherwise) can be a good t= hing too. - Jordan=20 Sent from my iPad On Jul 9, 2013, at 7:39, John Baldwin <jhb@freebsd.org> wrote: > I'll only speak to the GC point. For anonymous shm's (SHM_ANON), the segm= ent > is in fact GC'd on last close as there is no name to hold a reference to i= t.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CC5D1844-EA9A-4B88-A842-F7F4307458ED>