Sjur Brændeland
2013-Mar-11 08:55 UTC
Opposite return values in vringh_notify_enable_kern() and virtqueue_enable_cb().
Hi Rusty, The two similar functions in vringh and virtqueue for turning on interrupts has opposite return values if there are buffers available in the ring. I think it would be better if these two functions aligned the use of return values. Maybe it's just me, but I got the logic for re-scheduling NAPI wrong due to this. /** * vringh_notify_enable_kern - we want to know if something changes. * @vrh: the vring. * * This always enables notifications, but returns true if there are * now more buffers available in the vring. */ bool vringh_notify_enable_kern(struct vringh *vrh) /** * virtqueue_enable_cb - restart callbacks after disable_cb. * @vq: the struct virtqueue we're talking about. * * This re-enables callbacks; it returns "false" if there are pending * buffers in the queue, .... */ bool virtqueue_enable_cb(struct virtqueue *_vq) Regards. Sjur
Rusty Russell
2013-Mar-12 04:32 UTC
Opposite return values in vringh_notify_enable_kern() and virtqueue_enable_cb().
Sjur Br?ndeland <sjurbren at gmail.com> writes:> Hi Rusty, > > The two similar functions in vringh and virtqueue for turning on > interrupts has opposite return values if there are buffers available > in the ring. I think it would be better if these two functions aligned > the use of return values. Maybe it's just me, but I got the logic > for re-scheduling NAPI wrong due to this.Wow. Firstly, the author of the original was an idiot for getting the API wrong. Secondly, the author of the second was an idiot for making it different. So I'm doubly an idiot. If I'd hit that I would have been far less polite :) Hmm, wait, vhost has them inverted, so maybe I can blame MST... anyway, I fixed that too. Thanks! Rusty.> /** > * vringh_notify_enable_kern - we want to know if something changes. > * @vrh: the vring. > * > * This always enables notifications, but returns true if there are > * now more buffers available in the vring. > */ > bool vringh_notify_enable_kern(struct vringh *vrh) > > /** > * virtqueue_enable_cb - restart callbacks after disable_cb. > * @vq: the struct virtqueue we're talking about. > * > * This re-enables callbacks; it returns "false" if there are pending > * buffers in the queue, .... > */ > bool virtqueue_enable_cb(struct virtqueue *_vq) > > Regards. > Sjur
Possibly Parallel Threads
- Opposite return values in vringh_notify_enable_kern() and virtqueue_enable_cb().
- [PATCH vringh 0/2] Introduce CAIF Virtio driver
- [PATCH vringh 0/2] Introduce CAIF Virtio driver
- [PATCHv2 vringh 0/3] Introduce CAIF Virtio driver
- [PATCHv2 vringh 0/3] Introduce CAIF Virtio driver