Date: Sun, 14 Mar 2004 00:53:55 -0500 (EST) From: Daniel Eischen <eischen@vigrid.com> To: Tim Robbins <tjr@freebsd.org> Cc: threads@freebsd.org Subject: Re: RFC: getc() and putc() as macros Message-ID: <Pine.GSO.4.10.10403140042001.29402-100000@pcnet5.pcnet.com> In-Reply-To: <20040314010805.GA21447@cat.robbins.dropbear.id.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 14 Mar 2004, Tim Robbins wrote: > On Sat, Mar 13, 2004 at 10:05:14AM -0500, Daniel Eischen wrote: > > > On Sat, 13 Mar 2004, Tim Robbins wrote: > > > > > The patch below re-adds macro versions of getc(), getchar(), putc(), > > > putchar(), feof(), ferror(), fileno() and clearerr(), using the value of > > > __isthreaded to decide between the fast inline single-threaded code and > > > the more general function equivalent (as suggested by Alfred). Is this > > > approach safe? > > > > I don't really like this. It exposes __isthreaded and others > > that are implementation. > > Can you think of a better way? I think it was I that got rid of the macros for getc() et al. I did it when libc_r was divorced from libc, and the macro _THREAD_SAFE was no longer necessary. Solaris uses _REENTRANT to toggle between macros and functions. For the macro versions, it accesses the FILE directly instead of making a function call. I think the _unlocked versions of the functions are there for a reason. If an application isn't going to be threaded, then it can always use the unlocked versions... -- Dan Eischen
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.4.10.10403140042001.29402-100000>