| Message ID | 20230830045722.611224-2-fengli@smartx.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Fix vhost reconnect issues | expand |
> On Aug 30, 2023, at 12:57 AM, Li Feng <fengli@smartx.com> wrote: > > When the vhost-user is reconnecting to the backend, and if the vhost-user fails > at the get_features in vhost_dev_init(), then the reconnect will fail > and it will not be retriggered forever. > > The reason is: > When the vhost-user fails at get_features, the vhost_dev_cleanup will be called > immediately. > > vhost_dev_cleanup calls 'memset(hdev, 0, sizeof(struct vhost_dev))'. > > The reconnect path is: > vhost_user_blk_event > vhost_user_async_close(.. vhost_user_blk_disconnect ..) > qemu_chr_fe_set_handlers <----- clear the notifier callback > schedule vhost_user_async_close_bh > > The vhost->vdev is null, so the vhost_user_blk_disconnect will not be > called, then the event fd callback will not be reinstalled. > > All vhost-user devices have this issue, including vhost-user-blk/scsi. > > With this patch, if the vdev->vdev is null, the fd callback will still > be reinstalled. > > Fixes: 71e076a07d ("hw/virtio: generalise CHR_EVENT_CLOSED handling") > Reviewed-by: Raphael Norwitz <raphael.norwitz@nutanix.com> > Signed-off-by: Li Feng <fengli@smartx.com> > --- > hw/block/vhost-user-blk.c | 2 +- > hw/scsi/vhost-user-scsi.c | 3 ++- > hw/virtio/vhost-user-gpio.c | 2 +- > hw/virtio/vhost-user.c | 9 +++++++-- > include/hw/virtio/vhost-user.h | 3 ++- > 5 files changed, 13 insertions(+), 6 deletions(-) > > diff --git a/hw/block/vhost-user-blk.c b/hw/block/vhost-user-blk.c > index 3c69fa47d5..95c758200d 100644 > --- a/hw/block/vhost-user-blk.c > +++ b/hw/block/vhost-user-blk.c > @@ -391,7 +391,7 @@ static void vhost_user_blk_event(void *opaque, QEMUChrEvent event) > case CHR_EVENT_CLOSED: > /* defer close until later to avoid circular close */ > vhost_user_async_close(dev, &s->chardev, &s->dev, > - vhost_user_blk_disconnect); > + vhost_user_blk_disconnect, vhost_user_blk_event); > break; > case CHR_EVENT_BREAK: > case CHR_EVENT_MUX_IN: > diff --git a/hw/scsi/vhost-user-scsi.c b/hw/scsi/vhost-user-scsi.c > index a7fa8e8df2..e931df9f5b 100644 > --- a/hw/scsi/vhost-user-scsi.c > +++ b/hw/scsi/vhost-user-scsi.c > @@ -236,7 +236,8 @@ static void vhost_user_scsi_event(void *opaque, QEMUChrEvent event) > case CHR_EVENT_CLOSED: > /* defer close until later to avoid circular close */ > vhost_user_async_close(dev, &vs->conf.chardev, &vsc->dev, > - vhost_user_scsi_disconnect); > + vhost_user_scsi_disconnect, > + vhost_user_scsi_event); > break; > case CHR_EVENT_BREAK: > case CHR_EVENT_MUX_IN: > diff --git a/hw/virtio/vhost-user-gpio.c b/hw/virtio/vhost-user-gpio.c > index d9979aa5db..04c2cc79f4 100644 > --- a/hw/virtio/vhost-user-gpio.c > +++ b/hw/virtio/vhost-user-gpio.c > @@ -283,7 +283,7 @@ static void vu_gpio_event(void *opaque, QEMUChrEvent event) > case CHR_EVENT_CLOSED: > /* defer close until later to avoid circular close */ > vhost_user_async_close(dev, &gpio->chardev, &gpio->vhost_dev, > - vu_gpio_disconnect); > + vu_gpio_disconnect, vu_gpio_event); > break; > case CHR_EVENT_BREAK: > case CHR_EVENT_MUX_IN: > diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c > index 8dcf049d42..12c4a41f30 100644 > --- a/hw/virtio/vhost-user.c > +++ b/hw/virtio/vhost-user.c > @@ -2643,6 +2643,7 @@ typedef struct { > DeviceState *dev; > CharBackend *cd; > struct vhost_dev *vhost; > + IOEventHandler *event_cb; > } VhostAsyncCallback; > > static void vhost_user_async_close_bh(void *opaque) > @@ -2657,7 +2658,10 @@ static void vhost_user_async_close_bh(void *opaque) > */ > if (vhost->vdev) { > data->cb(data->dev); > - } > + } else if (data->event_cb) { > + qemu_chr_fe_set_handlers(data->cd, NULL, NULL, data->event_cb, > + NULL, data->dev, NULL, true); > + } > > g_free(data); > } > @@ -2669,7 +2673,8 @@ static void vhost_user_async_close_bh(void *opaque) > */ > void vhost_user_async_close(DeviceState *d, > CharBackend *chardev, struct vhost_dev *vhost, > - vu_async_close_fn cb) > + vu_async_close_fn cb, > + IOEventHandler *event_cb) > { > if (!runstate_check(RUN_STATE_SHUTDOWN)) { > /* > diff --git a/include/hw/virtio/vhost-user.h b/include/hw/virtio/vhost-user.h > index 191216a74f..649e9dd54f 100644 > --- a/include/hw/virtio/vhost-user.h > +++ b/include/hw/virtio/vhost-user.h > @@ -84,6 +84,7 @@ typedef void (*vu_async_close_fn)(DeviceState *cb); > > void vhost_user_async_close(DeviceState *d, > CharBackend *chardev, struct vhost_dev *vhost, > - vu_async_close_fn cb); > + vu_async_close_fn cb, > + IOEventHandler *event_cb); > > #endif > -- > 2.41.0 > >
diff --git a/hw/block/vhost-user-blk.c b/hw/block/vhost-user-blk.c index 3c69fa47d5..95c758200d 100644 --- a/hw/block/vhost-user-blk.c +++ b/hw/block/vhost-user-blk.c @@ -391,7 +391,7 @@ static void vhost_user_blk_event(void *opaque, QEMUChrEvent event) case CHR_EVENT_CLOSED: /* defer close until later to avoid circular close */ vhost_user_async_close(dev, &s->chardev, &s->dev, - vhost_user_blk_disconnect); + vhost_user_blk_disconnect, vhost_user_blk_event); break; case CHR_EVENT_BREAK: case CHR_EVENT_MUX_IN: diff --git a/hw/scsi/vhost-user-scsi.c b/hw/scsi/vhost-user-scsi.c index a7fa8e8df2..e931df9f5b 100644 --- a/hw/scsi/vhost-user-scsi.c +++ b/hw/scsi/vhost-user-scsi.c @@ -236,7 +236,8 @@ static void vhost_user_scsi_event(void *opaque, QEMUChrEvent event) case CHR_EVENT_CLOSED: /* defer close until later to avoid circular close */ vhost_user_async_close(dev, &vs->conf.chardev, &vsc->dev, - vhost_user_scsi_disconnect); + vhost_user_scsi_disconnect, + vhost_user_scsi_event); break; case CHR_EVENT_BREAK: case CHR_EVENT_MUX_IN: diff --git a/hw/virtio/vhost-user-gpio.c b/hw/virtio/vhost-user-gpio.c index d9979aa5db..04c2cc79f4 100644 --- a/hw/virtio/vhost-user-gpio.c +++ b/hw/virtio/vhost-user-gpio.c @@ -283,7 +283,7 @@ static void vu_gpio_event(void *opaque, QEMUChrEvent event) case CHR_EVENT_CLOSED: /* defer close until later to avoid circular close */ vhost_user_async_close(dev, &gpio->chardev, &gpio->vhost_dev, - vu_gpio_disconnect); + vu_gpio_disconnect, vu_gpio_event); break; case CHR_EVENT_BREAK: case CHR_EVENT_MUX_IN: diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c index 8dcf049d42..12c4a41f30 100644 --- a/hw/virtio/vhost-user.c +++ b/hw/virtio/vhost-user.c @@ -2643,6 +2643,7 @@ typedef struct { DeviceState *dev; CharBackend *cd; struct vhost_dev *vhost; + IOEventHandler *event_cb; } VhostAsyncCallback; static void vhost_user_async_close_bh(void *opaque) @@ -2657,7 +2658,10 @@ static void vhost_user_async_close_bh(void *opaque) */ if (vhost->vdev) { data->cb(data->dev); - } + } else if (data->event_cb) { + qemu_chr_fe_set_handlers(data->cd, NULL, NULL, data->event_cb, + NULL, data->dev, NULL, true); + } g_free(data); } @@ -2669,7 +2673,8 @@ static void vhost_user_async_close_bh(void *opaque) */ void vhost_user_async_close(DeviceState *d, CharBackend *chardev, struct vhost_dev *vhost, - vu_async_close_fn cb) + vu_async_close_fn cb, + IOEventHandler *event_cb) { if (!runstate_check(RUN_STATE_SHUTDOWN)) { /* diff --git a/include/hw/virtio/vhost-user.h b/include/hw/virtio/vhost-user.h index 191216a74f..649e9dd54f 100644 --- a/include/hw/virtio/vhost-user.h +++ b/include/hw/virtio/vhost-user.h @@ -84,6 +84,7 @@ typedef void (*vu_async_close_fn)(DeviceState *cb); void vhost_user_async_close(DeviceState *d, CharBackend *chardev, struct vhost_dev *vhost, - vu_async_close_fn cb); + vu_async_close_fn cb, + IOEventHandler *event_cb); #endif
When the vhost-user is reconnecting to the backend, and if the vhost-user fails at the get_features in vhost_dev_init(), then the reconnect will fail and it will not be retriggered forever. The reason is: When the vhost-user fails at get_features, the vhost_dev_cleanup will be called immediately. vhost_dev_cleanup calls 'memset(hdev, 0, sizeof(struct vhost_dev))'. The reconnect path is: vhost_user_blk_event vhost_user_async_close(.. vhost_user_blk_disconnect ..) qemu_chr_fe_set_handlers <----- clear the notifier callback schedule vhost_user_async_close_bh The vhost->vdev is null, so the vhost_user_blk_disconnect will not be called, then the event fd callback will not be reinstalled. All vhost-user devices have this issue, including vhost-user-blk/scsi. With this patch, if the vdev->vdev is null, the fd callback will still be reinstalled. Fixes: 71e076a07d ("hw/virtio: generalise CHR_EVENT_CLOSED handling") Signed-off-by: Li Feng <fengli@smartx.com> --- hw/block/vhost-user-blk.c | 2 +- hw/scsi/vhost-user-scsi.c | 3 ++- hw/virtio/vhost-user-gpio.c | 2 +- hw/virtio/vhost-user.c | 9 +++++++-- include/hw/virtio/vhost-user.h | 3 ++- 5 files changed, 13 insertions(+), 6 deletions(-)