Date: Sun, 17 Jan 2016 09:26:42 +0100 From: Mateusz Guzik <mjguzik@gmail.com> To: Chagin Dmitry <dchagin@freebsd.org> Cc: Konstantin Belousov <kib@kib.kiev.ua>, Vijay Singh <vijju.singh@gmail.com>, freebsd-hackers@freebsd.org Subject: Re: irrelevant locking Message-ID: <20160117082642.GC1963@dft-labs.eu> In-Reply-To: <20160117080515.GA32432@chd.heemeyer.club> References: <20160116195819.GA41610@chd.heemeyer.club> <20160116202643.GL3942@kib.kiev.ua> <CALCNsJT_gH5gJaB%2ByVQRcON84JntSUevG8-X-0Z5_13DkPC%2BBg@mail.gmail.com> <20160116224312.GA1963@dft-labs.eu> <20160117034315.GN3942@kib.kiev.ua> <20160117080515.GA32432@chd.heemeyer.club>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jan 17, 2016 at 11:05:15AM +0300, Chagin Dmitry wrote: > On Sun, Jan 17, 2016 at 05:43:15AM +0200, Konstantin Belousov wrote: > > On Sat, Jan 16, 2016 at 11:43:13PM +0100, Mateusz Guzik wrote: > > > The real question is if it would make sense to add the bit to elf aux > > > vector to save the call as done by the loader. > > I once did a pass to remove (most of) sysctls executed during process > > startup. issetugid indeed may be treated same. > > like a Linux AT_SECURE? our P_SUGID bit can be changed after exec, > so the result of issetugid() implementated on auxv table will be unnafected > by calls to setuid(), setgid() or other such calls. > This is only to save issetugid call done early by the loader and it is perfectly safe at that point. This definitely is not a replacement for the syscall in general. -- Mateusz Guzik <mjguzik gmail.com>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20160117082642.GC1963>