Date: Mon, 21 Jul 2008 09:39:16 +0200 From: "Antoine Brodin" <antoine@FreeBSD.org> To: "Garrett Cooper" <yanefbsd@gmail.com> Cc: freebsd-bugs@freebsd.org Subject: Re: bin/125680: atacontrol(8): atacontrol depends on executable in /usr Message-ID: <f19c444a0807210039v99e8701g6c9c1fcb4f3e7b5@mail.gmail.com> In-Reply-To: <7d6fde3d0807201714g49eb4a80ncfcc1cc800ad595e@mail.gmail.com> References: <200807202210.m6KMA4cm032331@freefall.freebsd.org> <7d6fde3d0807201714g49eb4a80ncfcc1cc800ad595e@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jul 21, 2008 at 2:14 AM, Garrett Cooper <yanefbsd@gmail.com> wrote: > Good catch. C apps shouldn't depend on other commands' existence for > security and performance reasons. > > A few comments: > 1. Why is forking another process necessary? In your patch above > you're forcing the parent to do the work while the child goes and > dies, so there really isn't any benefit to forking at all other than > just exercise fork a bit. > 2. If you're going to fork another process, wouldn't it be wiser to > waitpid(2) until the child's done? Hi, In the patch, the child does the work and the parent dies. The "/usr/bin/nice -n 20 /bin/dd if=/dev/ar%d of=/dev/null bs=1m &" was launched in the background (notice the "&"). Cheers, Antoine
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?f19c444a0807210039v99e8701g6c9c1fcb4f3e7b5>