linux-kernel - Re: [PATCH v2] scsi: virtio_scsi: Fix poential NULL pointer dereference in virtscsi_rescan

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 14 Feb 2023 10:36:56 +0800
From:   Zheng Hacker <hackerzheng666@...il.com>
To:     "Michael S. Tsirkin" <mst@...hat.com>
Cc:     Zheng Wang <zyytlz.wz@....com>, jasowang@...hat.com,
        pbonzini@...hat.com, stefanha@...hat.com, jejb@...ux.ibm.com,
        martin.petersen@...cle.com,
        virtualization@...ts.linux-foundation.org,
        linux-scsi@...r.kernel.org, linux-kernel@...r.kernel.org,
        security@...nel.org, alex000young@...il.com
Subject: Re: [PATCH v2] scsi: virtio_scsi: Fix poential NULL pointer
 dereference in virtscsi_rescan_hotunplug

Michael S. Tsirkin <mst@...hat.com> 于2023年2月13日周一 20:09写道：
>
> On Thu, Feb 02, 2023 at 02:41:24PM +0800, Zheng Wang wrote:
> > There is no check about the return value of kmalloc in
> > virtscsi_rescan_hotunplug. Add the check to avoid use
> > of null pointer 'inq_result' in case of the failure
> > of kmalloc.
> >
> > Signed-off-by: Zheng Wang <zyytlz.wz@....com>
> > ---
>
> I fixed a typo in subject and tweaked the patch a bit

Thanks for your kindness ^^

>
> > v2:
> > - add kfree to avoid memory leak
> > ---
> >  drivers/scsi/virtio_scsi.c | 13 +++++++++++--
> >  1 file changed, 11 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
> > index d07d24c06b54..a66d8815d738 100644
> > --- a/drivers/scsi/virtio_scsi.c
> > +++ b/drivers/scsi/virtio_scsi.c
> > @@ -330,7 +330,7 @@ static void virtscsi_handle_param_change(struct virtio_scsi *vscsi,
> >       scsi_device_put(sdev);
> >  }
> >
> > -static void virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
> > +static int virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
> >  {
> >       struct scsi_device *sdev;
> >       struct Scsi_Host *shost = virtio_scsi_host(vscsi->vdev);
> > @@ -338,6 +338,11 @@ static void virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
> >       int result, inquiry_len, inq_result_len = 256;
> >       char *inq_result = kmalloc(inq_result_len, GFP_KERNEL);
> >
> > +     if (!inq_result) {
> > +             kfree(inq_result);
> > +             return -ENOMEM;
> > +     }
> > +
> >       shost_for_each_device(sdev, shost) {
> >               inquiry_len = sdev->inquiry_len ? sdev->inquiry_len : 36;
> >
> > @@ -366,6 +371,7 @@ static void virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
> >       }
> >
> >       kfree(inq_result);
> > +     return 0;
> >  }
> >
> >  static void virtscsi_handle_event(struct work_struct *work)
> > @@ -374,12 +380,15 @@ static void virtscsi_handle_event(struct work_struct *work)
> >               container_of(work, struct virtio_scsi_event_node, work);
> >       struct virtio_scsi *vscsi = event_node->vscsi;
> >       struct virtio_scsi_event *event = &event_node->event;
> > +     int ret = 0;
> >
>
> dropped = 0 here

Get it.

>
> >       if (event->event &
> >           cpu_to_virtio32(vscsi->vdev, VIRTIO_SCSI_T_EVENTS_MISSED)) {
>
> and moved declaration here.

Will do in the next version of patch.
>
> >               event->event &= ~cpu_to_virtio32(vscsi->vdev,
> >                                                  VIRTIO_SCSI_T_EVENTS_MISSED);
> > -             virtscsi_rescan_hotunplug(vscsi);
> > +             ret = virtscsi_rescan_hotunplug(vscsi);
> > +             if (ret)
> > +                     return;
> >               scsi_scan_host(virtio_scsi_host(vscsi->vdev));
> >       }
> >
> > --
> > 2.25.1
> >
> >
>