Re: [Qemu-devel] [PATCH v6 03/10] virtio-blk: fix reference a pointer which might be freed

[Michael S. Tsirkin]

On Thu, Aug 14, 2014 at 02:39:22PM +0400, Michael Tokarev wrote:
> 14.08.2014 14:37, Michael S. Tsirkin wrote:
> > On Thu, Aug 14, 2014 at 03:29:14PM +0800, zhanghailiang wrote:
> >> In function virtio_blk_handle_request, it may freed memory pointed by req,
> >> So do not access member of req after calling this function.
> >>
> >> Reviewed-by: Stefan Hajnoczi <address@hidden>
> >> Signed-off-by: zhanghailiang <address@hidden>
> > 
> > Reviewed-by: Michael S. Tsirkin <address@hidden>
> 
> Just a small nitpick...
> 
> > Stefan want to pick up this one?
> > 
> >> ---
> >>  hw/block/virtio-blk.c | 5 +++--
> >>  1 file changed, 3 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
> >> index c241c50..54a853a 100644
> >> --- a/hw/block/virtio-blk.c
> >> +++ b/hw/block/virtio-blk.c
> >> @@ -458,7 +458,7 @@ static void virtio_blk_handle_output(VirtIODevice 
> >> *vdev, VirtQueue *vq)
> >>  static void virtio_blk_dma_restart_bh(void *opaque)
> >>  {
> >>      VirtIOBlock *s = opaque;
> >> -    VirtIOBlockReq *req = s->rq;
> >> +    VirtIOBlockReq *req = s->rq, *next = NULL;
> 
> There's no need to initialize next here.
> With init like this I for one instinctively start searching below
> how this `NULL' is used (which it isn't).
> 
> I'd declare it inside the loop too, since it is a local-to-loop
> var, but that doesn't matter much.
> 
> /mjt

Good point, I agree.

> >>      MultiReqBuffer mrb = {
> >>          .num_writes = 0,
> >>      };
> >> @@ -469,8 +469,9 @@ static void virtio_blk_dma_restart_bh(void *opaque)
> >>      s->rq = NULL;
> >>  
> >>      while (req) {
> >> +        next = req->next;
> >>          virtio_blk_handle_request(req, &mrb);
> >> -        req = req->next;
> >> +        req = next;
> >>      }
> >>  
> >>      virtio_submit_multiwrite(s->bs, &mrb);
> >> -- 
> >> 1.7.12.4
> >>