Linux Virtualization - Jan 2008 - [PATCH] Use memset(0) instead of setting cb_mode explicitly

hrtimer's only have a cb_mode if CONFIG_HIGH_RES_TIMER is enabled.  This
patch
memset()s the tx_timer structure to 0 which has the same effect as setting
cb_mode to HRTIMER_CB_SOFTIRQ without requiring CONFIG_HIGH_RES_TIMER to be
set.  This seems to be what other drivers do.

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 47e4fce..d11a670 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -434,9 +434,9 @@ static int virtnet_probe(struct virtio_device *vdev)
 	netif_napi_add(dev, &vi->napi, virtnet_poll, napi_weight);
 	vi->dev = dev;
 	vi->vdev = vdev;
+	memset(&vi->tx_timer, 0, sizeof(vi->tx_timer));
 	hrtimer_init(&vi->tx_timer, CLOCK_REALTIME, HRTIMER_MODE_REL);
 	vi->tx_timer.function = kick_xmit;
-	vi->tx_timer.cb_mode = HRTIMER_CB_SOFTIRQ;
 	vi->out_max = -1U;
 
 	tasklet_init(&vi->xmit_free, xmit_free, (unsigned long)vi);

On Friday 04 January 2008 02:57:14 Anthony Liguori
wrote:
> hrtimer's only have a cb_mode if CONFIG_HIGH_RES_TIMER is enabled. 
This
> patch memset()s the tx_timer structure to 0 which has the same effect as
> setting cb_mode to HRTIMER_CB_SOFTIRQ without requiring
> CONFIG_HIGH_RES_TIMER to be set.  This seems to be what other drivers do.
Thanks, applied.

Rusty.