Date: Mon, 24 Mar 2003 11:07:21 -0600 From: "Jacques A. Vidrine" <nectar@FreeBSD.org> To: Martin Blapp <mb@imp.ch> Cc: Martin Blapp <mbr@FreeBSD.org>, src-committers@FreeBSD.org, cvs-src@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/contrib/isc-dhcp/client dhclient.c Message-ID: <20030324170721.GD1911@madman.celabo.org> In-Reply-To: <20030324170856.O21003@cvs.imp.ch> References: <200303232329.h2NNTaI5089971@repoman.freebsd.org> <20030324160604.GB1911@madman.celabo.org> <20030324170856.O21003@cvs.imp.ch>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Mar 24, 2003 at 05:35:49PM +0100, Martin Blapp wrote: > > Hi, > > > Are you sure that all values of pid_t can be represented in a signed > > int? > > sys/alpha/include/ansi.h:#define _BSD_PID_T_ int /* process [group] */ > sys/i386/include/ansi.h:#define _BSD_PID_T_ int /* process [group] */ > sys/sys/types.h:typedef _BSD_PID_T_ pid_t; /* process id */ So we hope that never changes? :-) On -CURRENT, apparently pid_t is always int32_t. But, %d is `int' and I'm not sure that `int' is always 32 bits and will be forever. > > + pid_t oldpid; > > ! e = fscanf(pidfd, "%d", &oldpid); > > > > I'd probably opt to do the fscanf with an unsigned long, and then cast > > the value to (pid_t). *shrug* > > or use a unsigned int for it ? Will we have to change this for later ? Why int rather than long? When doing I/O, use the widest standard types available and then see if the result fits in your target. (Don't ask me how to check the range in portable way :-) Cheers, -- Jacques A. Vidrine <nectar@celabo.org> http://www.celabo.org/ NTT/Verio SME . FreeBSD UNIX . Heimdal Kerberos jvidrine@verio.net . nectar@FreeBSD.org . nectar@kth.se To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030324170721.GD1911>