Re: [PATCH v2] virtio-scsi: fix race in virtio_scsi_dataplane_start()

2022-08-09 Thread Emanuele Giuseppe Esposito
Am 08/08/2022 um 18:21 schrieb Stefan Hajnoczi: > As soon as virtio_scsi_data_plane_start() attaches host notifiers the > IOThread may start virtqueue processing. There is a race between > IOThread virtqueue processing and virtio_scsi_data_plane_start() because > it only assigns s->dataplane_sta

Re: [PATCH v2] virtio-scsi: fix race in virtio_scsi_dataplane_start()

2022-08-08 Thread Michael S. Tsirkin
On Mon, Aug 08, 2022 at 12:21:34PM -0400, Stefan Hajnoczi wrote: > As soon as virtio_scsi_data_plane_start() attaches host notifiers the > IOThread may start virtqueue processing. There is a race between > IOThread virtqueue processing and virtio_scsi_data_plane_start() because > it only assigns s-

[PATCH v2] virtio-scsi: fix race in virtio_scsi_dataplane_start()

2022-08-08 Thread Stefan Hajnoczi
As soon as virtio_scsi_data_plane_start() attaches host notifiers the IOThread may start virtqueue processing. There is a race between IOThread virtqueue processing and virtio_scsi_data_plane_start() because it only assigns s->dataplane_started after attaching host notifiers. When a virtqueue hand