On Mon, Mar 25, 2013 at 01:10:33PM +0200, Michael S. Tsirkin wrote:
> On Mon, Mar 25, 2013 at 03:48:06PM +0800, Asias He wrote:
> > On Sun, Mar 24, 2013 at 05:20:09PM +0200, Michael S. Tsirkin wrote:
> > > On Fri, Mar 22, 2013 at 01:39:05PM +0800, Asias He wrote:
> > > > In commit 365a7150094 ([SCSI] virtio-scsi: hotplug support for
> > > > virtio-scsi), hotplug support is added to virtio-scsi.
> > > >
> > > > This patch adds hotplug and hotunplug support to tcm_vhost.
> > > >
> > > > You can create or delete a LUN in targetcli to hotplug or hotunplug a
> > > > LUN in guest.
> > > >
> > > > Changes in v4:
> > > > - Drop tcm_vhost_check_endpoint in tcm_vhost_send_evt
> > > > - Add tcm_vhost_check_endpoint in vhost_scsi_evt_handle_kick
> > > >
> > > > Changes in v3:
> > > > - Separate the bug fix to another thread
> > > >
> > > > Changes in v2:
> > > > - Remove code duplication in tcm_vhost_{hotplug,hotunplug}
> > > > - Fix racing of vs_events_nr
> > > > - Add flush fix patch to this series
> > > >
> > > > Signed-off-by: Asias He <as...@redhat.com>
> > > > Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
> > > > ---
> > > > drivers/vhost/tcm_vhost.c | 212
> > > > ++++++++++++++++++++++++++++++++++++++++++++--
> > > > drivers/vhost/tcm_vhost.h | 10 +++
> > > > 2 files changed, 217 insertions(+), 5 deletions(-)
> > > >
> > > > diff --git a/drivers/vhost/tcm_vhost.c b/drivers/vhost/tcm_vhost.c
> > > > index d81e3a9..e734ead 100644
> > > > --- a/drivers/vhost/tcm_vhost.c
> > > > +++ b/drivers/vhost/tcm_vhost.c
> > > > @@ -62,6 +62,9 @@ enum {
> > > >
> > > > #define VHOST_SCSI_MAX_TARGET 256
> > > > #define VHOST_SCSI_MAX_VQ 128
> > > > +#define VHOST_SCSI_MAX_EVENT 128
> > > > +
> > > > +#define VHOST_SCSI_FEATURES (VHOST_FEATURES | (1ULL <<
> > > > VIRTIO_SCSI_F_HOTPLUG))
> > > >
> > > > struct vhost_scsi {
> > > > /* Protected by vhost_scsi->dev.mutex */
> > > > @@ -73,6 +76,12 @@ struct vhost_scsi {
> > > >
> > > > struct vhost_work vs_completion_work; /* cmd completion work
> > > > item */
> > > > struct llist_head vs_completion_list; /* cmd completion queue */
> > > > +
> > > > + struct vhost_work vs_event_work; /* evt injection work item */
> > > > + struct llist_head vs_event_list; /* evt injection queue */
> > > > +
> > > > + bool vs_events_dropped; /* any missed events, protected by
> > > > dev.mutex */
> > > > + u64 vs_events_nr; /* num of pending events, protected by
> > > > dev.mutex */
> > >
> > > Woa u64. We allow up to 128 of these. int will do.
> >
> > u64 is used before we limit the total number of events, changed to int.
> >
> > > > };
> > > >
> > > > /* Local pointer to allocated TCM configfs fabric module */
> > > > @@ -120,6 +129,16 @@ static bool tcm_vhost_check_endpoint(struct
> > > > vhost_virtqueue *vq)
> > > > return ret;
> > > > }
> > > >
> > > > +static bool tcm_vhost_check_events_dropped(struct vhost_scsi *vs)
> > > > +{
> > > > + bool ret;
> > > > +
> > > > + mutex_lock(&vs->dev.mutex);
> > > > + ret = vs->vs_events_dropped;
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > +
> > > > + return ret;
> > > > +}
> > > > static int tcm_vhost_check_true(struct se_portal_group *se_tpg)
> > > > {
> > > > return 1;
> > > > @@ -370,6 +389,36 @@ static int tcm_vhost_queue_tm_rsp(struct se_cmd
> > > > *se_cmd)
> > > > return 0;
> > > > }
> > > >
> > > > +static void tcm_vhost_free_evt(struct vhost_scsi *vs, struct
> > > > tcm_vhost_evt *evt)
> > > > +{
> > > > + mutex_lock(&vs->dev.mutex);
> > > > + vs->vs_events_nr--;
> > > > + kfree(evt);
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > +}
> > > > +
> > > > +static struct tcm_vhost_evt *tcm_vhost_allocate_evt(struct vhost_scsi
> > > > *vs,
> > > > + u32 event, u32 reason)
> > > > +{
> > > > + struct tcm_vhost_evt *evt;
> > > > +
> > > > + mutex_lock(&vs->dev.mutex);
> > > > + if (vs->vs_events_nr > VHOST_SCSI_MAX_EVENT) {
> > >
> > >
> > > Se eventfd dropped flag here and stop worrying about it in callers?
> >
> > Set the flag here now and do not bother the callers.
> >
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > + return NULL;
> > > > + }
> > > > +
> > > > + evt = kzalloc(sizeof(*evt), GFP_KERNEL);
> > > > + if (evt) {
> > > > + evt->event.event = event;
> > > > + evt->event.reason = reason;
> > > > + vs->vs_events_nr++;
> > > > + }
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > +
> > > > + return evt;
> > > > +}
> > > > +
> > > > static void vhost_scsi_free_cmd(struct tcm_vhost_cmd *tv_cmd)
> > > > {
> > > > struct se_cmd *se_cmd = &tv_cmd->tvc_se_cmd;
> > > > @@ -388,6 +437,77 @@ static void vhost_scsi_free_cmd(struct
> > > > tcm_vhost_cmd *tv_cmd)
> > > > kfree(tv_cmd);
> > > > }
> > > >
> > > > +static void tcm_vhost_do_evt_work(struct vhost_scsi *vs,
> > > > + struct virtio_scsi_event *event)
> > > > +{
> > > > + struct vhost_virtqueue *vq = &vs->vqs[VHOST_SCSI_VQ_EVT];
> > > > + struct virtio_scsi_event __user *eventp;
> > > > + unsigned out, in;
> > > > + int head, ret;
> > > > +
> > > > + if (!tcm_vhost_check_endpoint(vq))
> > > > + return;
> > > > +
> > > > + mutex_lock(&vq->mutex);
> > > > +again:
> > > > + vhost_disable_notify(&vs->dev, vq);
> > > > + head = vhost_get_vq_desc(&vs->dev, vq, vq->iov,
> > > > + ARRAY_SIZE(vq->iov), &out, &in,
> > > > + NULL, NULL);
> > > > + if (head < 0) {
> > > > + mutex_lock(&vs->dev.mutex);
> > >
> > > This nests dev mutex within vq mutex, can cause deadlock.
> > > Should be the other way around.
> >
> > Changed the order.
> >
> > > > + vs->vs_events_dropped = true;
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > + goto out;
> > > > + }
> > > > + if (head == vq->num) {
> > > > + if (vhost_enable_notify(&vs->dev, vq))
> > > > + goto again;
> > > > + mutex_lock(&vs->dev.mutex);
> > > > + vs->vs_events_dropped = true;
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > + goto out;
> > > > + }
> > > > +
> > > > + if ((vq->iov[out].iov_len != sizeof(struct virtio_scsi_event)))
> > > > {
> > > > + vq_err(vq, "Expecting virtio_scsi_event, got %zu
> > > > bytes\n",
> > > > + vq->iov[out].iov_len);
> > > > + goto out;
> > > > + }
> > > > +
> > > > + mutex_lock(&vs->dev.mutex);
> > > > + if (vs->vs_events_dropped) {
> > > > + event->event |= VIRTIO_SCSI_T_EVENTS_MISSED;
> > > > + vs->vs_events_dropped = false;
> > > > + }
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > +
> > > > + eventp = vq->iov[out].iov_base;
> > > > + ret = __copy_to_user(eventp, event, sizeof(*event));
> > >
> > > This assumes specific framing, please don't do this.
> >
> > As I mentioned in the previous mail about this issue. Let's do the
> > conversion to the no framing assumption mode in separate patch which
> > converts all of them. Specific framing is assumed everywhere currently.
>
> Yes this is a bug, let's fix it. I don't see any reason to add
> buggy code, then rework it.
Firstly, I agree with we need the conversion. But I do not think this is
a bug. The virtio-scsi guest side is always using the framing currently.
Until we modify the guest side code, this is not a bug.
This is just the matter of how we want to do the refactor.
> > > > + if (!ret)
> > > > + vhost_add_used_and_signal(&vs->dev, vq, head, 0);
> > > > + else
> > > > + vq_err(vq, "Faulted on tcm_vhost_send_event\n");
> > > > +out:
> > > > + mutex_unlock(&vq->mutex);
> > > > +}
> > > > +
> > > > +static void tcm_vhost_evt_work(struct vhost_work *work)
> > > > +{
> > > > + struct vhost_scsi *vs = container_of(work, struct vhost_scsi,
> > > > + vs_event_work);
> > > > + struct tcm_vhost_evt *evt;
> > > > + struct llist_node *llnode;
> > > > +
> > > > + llnode = llist_del_all(&vs->vs_event_list);
> > > > + while (llnode) {
> > > > + evt = llist_entry(llnode, struct tcm_vhost_evt, list);
> > > > + llnode = llist_next(llnode);
> > > > + tcm_vhost_do_evt_work(vs, &evt->event);
> > > > + tcm_vhost_free_evt(vs, evt);
> > > > + }
> > > > +}
> > > > +
> > > > /* Fill in status and signal that we are done processing this command
> > > > *
> > > > * This is scheduled in the vhost work queue so we are called with the
> > > > owner
> > > > @@ -785,9 +905,46 @@ static void vhost_scsi_ctl_handle_kick(struct
> > > > vhost_work *work)
> > > > pr_debug("%s: The handling func for control queue.\n",
> > > > __func__);
> > > > }
> > > >
> > > > +static int tcm_vhost_send_evt(struct vhost_scsi *vs, struct
> > > > tcm_vhost_tpg *tpg,
> > > > + struct se_lun *lun, u32 event, u32 reason)
> > > > +{
> > > > + struct tcm_vhost_evt *evt;
> > > > +
> > > > + evt = tcm_vhost_allocate_evt(vs, event, reason);
> > > > + if (!evt) {
> > > > + mutex_lock(&vs->dev.mutex);
> > > > + vs->vs_events_dropped = true;
> > > > + mutex_unlock(&vs->dev.mutex);
> > > > + return -ENOMEM;
> > > > + }
> > > > +
> > > > + if (tpg && lun) {
> > > > + /* TODO: share lun setup code with virtio-scsi.ko */
> > > > + evt->event.lun[0] = 0x01;
> > > > + evt->event.lun[1] = tpg->tport_tpgt & 0xFF;
> > > > + if (lun->unpacked_lun >= 256)
> > > > + evt->event.lun[2] = lun->unpacked_lun >> 8 |
> > > > 0x40 ;
> > > > + evt->event.lun[3] = lun->unpacked_lun & 0xFF;
> > > > + }
> > > > +
> > > > + llist_add(&evt->list, &vs->vs_event_list);
> > > > + vhost_work_queue(&vs->dev, &vs->vs_event_work);
> > > > +
> > > > + return 0;
> > > > +}
> > > > +
> > > > static void vhost_scsi_evt_handle_kick(struct vhost_work *work)
> > > > {
> > > > - pr_debug("%s: The handling func for event queue.\n", __func__);
> > > > + struct vhost_virtqueue *vq = container_of(work, struct
> > > > vhost_virtqueue,
> > > > + poll.work);
> > > > + struct vhost_scsi *vs = container_of(vq->dev, struct
> > > > vhost_scsi, dev);
> > > > +
> > > > + if (!tcm_vhost_check_endpoint(vq))
> > > > + return;
> > > > +
> > > > + if (tcm_vhost_check_events_dropped(vs))
> > > > + tcm_vhost_send_evt(vs, NULL, NULL,
> > > > VIRTIO_SCSI_T_NO_EVENT, 0);
> > > > +
> > > > }
> > > >
> > > > static void vhost_scsi_handle_kick(struct vhost_work *work)
> > > > @@ -844,6 +1001,7 @@ static int vhost_scsi_set_endpoint(
> > > > return -EEXIST;
> > > > }
> > > > tv_tpg->tv_tpg_vhost_count++;
> > > > + tv_tpg->vhost_scsi = vs;
> > > > vs->vs_tpg[tv_tpg->tport_tpgt] = tv_tpg;
> > > > smp_mb__after_atomic_inc();
> > > > match = true;
> > > > @@ -914,6 +1072,7 @@ static int vhost_scsi_clear_endpoint(
> > > > goto err_tpg;
> > > > }
> > > > tv_tpg->tv_tpg_vhost_count--;
> > > > + tv_tpg->vhost_scsi = NULL;
> > > > vs->vs_tpg[target] = NULL;
> > > > match = true;
> > > > mutex_unlock(&tv_tpg->tv_tpg_mutex);
> > > > @@ -947,6 +1106,9 @@ static int vhost_scsi_open(struct inode *inode,
> > > > struct file *f)
> > > > return -ENOMEM;
> > > >
> > > > vhost_work_init(&s->vs_completion_work,
> > > > vhost_scsi_complete_cmd_work);
> > > > + vhost_work_init(&s->vs_event_work, tcm_vhost_evt_work);
> > > > +
> > > > + s->vs_events_nr = 0;
> > > >
> > > > s->vqs[VHOST_SCSI_VQ_CTL].handle_kick =
> > > > vhost_scsi_ctl_handle_kick;
> > > > s->vqs[VHOST_SCSI_VQ_EVT].handle_kick =
> > > > vhost_scsi_evt_handle_kick;
> > > > @@ -989,11 +1151,12 @@ static void vhost_scsi_flush(struct vhost_scsi
> > > > *vs)
> > > > for (i = 0; i < VHOST_SCSI_MAX_VQ; i++)
> > > > vhost_scsi_flush_vq(vs, i);
> > > > vhost_work_flush(&vs->dev, &vs->vs_completion_work);
> > > > + vhost_work_flush(&vs->dev, &vs->vs_event_work);
> > > > }
> > > >
> > > > static int vhost_scsi_set_features(struct vhost_scsi *vs, u64 features)
> > > > {
> > > > - if (features & ~VHOST_FEATURES)
> > > > + if (features & ~VHOST_SCSI_FEATURES)
> > > > return -EOPNOTSUPP;
> > > >
> > > > mutex_lock(&vs->dev.mutex);
> > > > @@ -1039,7 +1202,7 @@ static long vhost_scsi_ioctl(struct file *f,
> > > > unsigned int ioctl,
> > > > return -EFAULT;
> > > > return 0;
> > > > case VHOST_GET_FEATURES:
> > > > - features = VHOST_FEATURES;
> > > > + features = VHOST_SCSI_FEATURES;
> > > > if (copy_to_user(featurep, &features, sizeof features))
> > > > return -EFAULT;
> > > > return 0;
> > > > @@ -1109,6 +1272,42 @@ static char *tcm_vhost_dump_proto_id(struct
> > > > tcm_vhost_tport *tport)
> > > > return "Unknown";
> > > > }
> > > >
> > > > +static int tcm_vhost_do_plug(struct tcm_vhost_tpg *tpg,
> > > > + struct se_lun *lun, bool plug)
> > > > +{
> > > > +
> > > > + struct vhost_scsi *vs = tpg->vhost_scsi;
> > > > + u32 reason;
> > > > +
> > > > + mutex_lock(&tpg->tv_tpg_mutex);
> > > > + vs = tpg->vhost_scsi;
> > > > + mutex_unlock(&tpg->tv_tpg_mutex);
> > > > + if (!vs)
> > > > + return -EOPNOTSUPP;
> > > > +
> > > > + if (!tcm_vhost_check_feature(vs, VIRTIO_SCSI_F_HOTPLUG))
> > > > + return -EOPNOTSUPP;
> > > > +
> > > > + if (plug)
> > > > + reason = VIRTIO_SCSI_EVT_RESET_RESCAN;
> > > > + else
> > > > + reason = VIRTIO_SCSI_EVT_RESET_REMOVED;
> > > > +
> > > > + return tcm_vhost_send_evt(vs, tpg, lun,
> > > > + VIRTIO_SCSI_T_TRANSPORT_RESET,
> > > > + reason);
> > > > +}
> > > > +
> > > > +static int tcm_vhost_hotplug(struct tcm_vhost_tpg *tpg, struct se_lun
> > > > *lun)
> > > > +{
> > > > + return tcm_vhost_do_plug(tpg, lun, true);
> > > > +}
> > > > +
> > > > +static int tcm_vhost_hotunplug(struct tcm_vhost_tpg *tpg, struct
> > > > se_lun *lun)
> > > > +{
> > > > + return tcm_vhost_do_plug(tpg, lun, false);
> > > > +}
> > > > +
> > > > static int tcm_vhost_port_link(struct se_portal_group *se_tpg,
> > > > struct se_lun *lun)
> > > > {
> > > > @@ -1119,18 +1318,21 @@ static int tcm_vhost_port_link(struct
> > > > se_portal_group *se_tpg,
> > > > tv_tpg->tv_tpg_port_count++;
> > > > mutex_unlock(&tv_tpg->tv_tpg_mutex);
> > > >
> > > > + tcm_vhost_hotplug(tv_tpg, lun);
> > > > +
> > > > return 0;
> > > > }
> > > >
> > > > static void tcm_vhost_port_unlink(struct se_portal_group *se_tpg,
> > > > - struct se_lun *se_lun)
> > > > + struct se_lun *lun)
> > > > {
> > > > struct tcm_vhost_tpg *tv_tpg = container_of(se_tpg,
> > > > struct tcm_vhost_tpg, se_tpg);
> > > > -
> > > > mutex_lock(&tv_tpg->tv_tpg_mutex);
> > > > tv_tpg->tv_tpg_port_count--;
> > > > mutex_unlock(&tv_tpg->tv_tpg_mutex);
> > > > +
> > > > + tcm_vhost_hotunplug(tv_tpg, lun);
> > > > }
> > > >
> > > > static struct se_node_acl *tcm_vhost_make_nodeacl(
> > > > diff --git a/drivers/vhost/tcm_vhost.h b/drivers/vhost/tcm_vhost.h
> > > > index 1d2ae7a..94e9ee53 100644
> > > > --- a/drivers/vhost/tcm_vhost.h
> > > > +++ b/drivers/vhost/tcm_vhost.h
> > > > @@ -53,6 +53,7 @@ struct tcm_vhost_nacl {
> > > > struct se_node_acl se_node_acl;
> > > > };
> > > >
> > > > +struct vhost_scsi;
> > > > struct tcm_vhost_tpg {
> > > > /* Vhost port target portal group tag for TCM */
> > > > u16 tport_tpgt;
> > > > @@ -70,6 +71,8 @@ struct tcm_vhost_tpg {
> > > > struct tcm_vhost_tport *tport;
> > > > /* Returned by tcm_vhost_make_tpg() */
> > > > struct se_portal_group se_tpg;
> > > > + /* Pointer back to struct vhost_scsi, protected by tv_tpg_mutex
> > > > */
> > > > + struct vhost_scsi *vhost_scsi;
> > > > };
> > > >
> > > > struct tcm_vhost_tport {
> > > > @@ -83,6 +86,13 @@ struct tcm_vhost_tport {
> > > > struct se_wwn tport_wwn;
> > > > };
> > > >
> > > > +struct tcm_vhost_evt {
> > > > + /* virtio_scsi event */
> > > > + struct virtio_scsi_event event;
> > > > + /* virtio_scsi event list, serviced from vhost worker thread */
> > > > + struct llist_node list;
> > > > +};
> > > > +
> > > > /*
> > > > * As per request from MST, keep TCM_VHOST related ioctl defines out of
> > > > * linux/vhost.h (user-space) for now..
> > > > --
> > > > 1.8.1.4
> >
> > --
> > Asias
--
Asias
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
