svn commit: r185356 - head/sys/dev/ixgbe

[M. Warner Losh]

In message: <20081126213204.14db9a63 at kan.dnsalias.net>
            Alexander Kabaev <kabaev at gmail.com> writes:
: On Thu, 27 Nov 2008 02:19:44 +0000 (UTC)
: Jack F Vogel <jfv at FreeBSD.org> wrote:
: 
: > Author: jfv
: > Date: Thu Nov 27 02:19:44 2008
: > New Revision: 185356
: > URL: http://svn.freebsd.org/changeset/base/185356
: > 
: > Log:
: >   Small nit I just noticed, a pre-decrement should be post.
: > 
: > Modified:
: >   head/sys/dev/ixgbe/ixgbe.c
: > 
: > Modified: head/sys/dev/ixgbe/ixgbe.c
: > ==============================================================================
: > --- head/sys/dev/ixgbe/ixgbe.c	Thu Nov 27 02:18:43 2008
: > (r185355) +++ head/sys/dev/ixgbe/ixgbe.c	Thu Nov 27 02:19:44
: > 2008	(r185356) @@ -3244,7 +3244,7 @@ fail:
: >  	 * the rings that completed, the failing case will have
: >  	 * cleaned up for itself. 'j' failed, so its the terminus.
: >  	 */
: > -	for (int i = 0; i < j; ++i) {
: > +	for (int i = 0; i < j; i++) {
: >  		rxr = &adapter->rx_rings[i];
: >  		for (int n = 0; n < adapter->num_rx_desc; n++) {
: >  			struct ixgbe_rx_buf *rxbuf;
: 
: Is C99 construct here intentional? If so, when did we agree on using
: only C99 compilers on our code base?

I'm not sure about *THIS* c99 construct, but we've been heavily
relying on the field name initializer stuff for a couple of years now.

eg:

static struct cdevsw lpt_cdevsw = {
	.d_version =	D_VERSION,
	.d_flags =	D_NEEDGIANT,
	.d_open =	lptopen,
	.d_close =	lptclose,
	.d_read =	lptread,
	.d_write =	lptwrite,
	.d_ioctl =	lptioctl,
	.d_name =	LPT_NAME,
};

Warner