On Thu, 27 May 2021 at 15:28, Gerd Hoffmann <kra...@redhat.com> wrote:
>
> From: Li Qiang <liq...@163.com>
>
> Call 'vugbm_buffer_destroy' in error path to avoid resource leak.
>
> Fixes: CVE-2021-3544
> Reported-by: Li Qiang <liq...@163.com>
> Reviewed-by: Prasad J Pandit <p...@fedoraproject.org>
> Signed-off-by: Li Qiang <liq...@163.com>
> Reviewed-by: Marc-André Lureau <marcandre.lur...@redhat.com>
> Message-Id: <20210516030403.107723-3-liq...@163.com>
> Signed-off-by: Gerd Hoffmann <kra...@redhat.com>
> ---
> contrib/vhost-user-gpu/vhost-user-gpu.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/contrib/vhost-user-gpu/vhost-user-gpu.c
> b/contrib/vhost-user-gpu/vhost-user-gpu.c
> index f73f292c9f72..b5e153d0d648 100644
> --- a/contrib/vhost-user-gpu/vhost-user-gpu.c
> +++ b/contrib/vhost-user-gpu/vhost-user-gpu.c
> @@ -349,6 +349,7 @@ vg_resource_create_2d(VuGpu *g,
> g_critical("%s: resource creation failed %d %d %d",
> __func__, c2d.resource_id, c2d.width, c2d.height);
> g_free(res);
> + vugbm_buffer_destroy(&res->buffer);
> cmd->error = VIRTIO_GPU_RESP_ERR_OUT_OF_MEMORY;
> return;
> }
Hi; Coverity reports this as a use-after-free: we free 'res'
and then on the next line we pass a pointer into this freed
memory to vugbm_buffer_destroy(), which dereferences it.
Probably the two lines should be in the other order ?
(CID 1453812)
thanks
-- PMM
