The "read" and "write" variables need to be signed for the
error
handling to work.
Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator")
Signed-off-by: Dan Carpenter <dan.carpenter at oracle.com>
---
drivers/vdpa/vdpa_sim/vdpa_sim.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c
index 6e8a0cf2fdeb..b3c800653056 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim.c
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c
@@ -132,7 +132,8 @@ static void vdpasim_work(struct work_struct *work)
vdpasim, work);
struct vdpasim_virtqueue *txq = &vdpasim->vqs[1];
struct vdpasim_virtqueue *rxq = &vdpasim->vqs[0];
- size_t read, write, total_write;
+ ssize_t read, write;
+ size_t total_write;
int err;
int pkts = 0;
--
2.25.1
Dan Carpenter
2020-Apr-06 14:45 UTC
[PATCH 2/2] vdpa: Fix pointer math bug in vdpasim_get_config()
If "offset" is non-zero then we end up copying from beyond the end of
the config because of pointer math. We can fix this by casting the
struct to a u8 pointer.
Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator")
Signed-off-by: Dan Carpenter <dan.carpenter at oracle.com>
---
Is it really worth letting people specify the offset?
drivers/vdpa/vdpa_sim/vdpa_sim.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c
index b3c800653056..e03c25765513 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim.c
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c
@@ -509,7 +509,7 @@ static void vdpasim_get_config(struct vdpa_device *vdpa,
unsigned int offset,
struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
if (offset + len < sizeof(struct virtio_net_config))
- memcpy(buf, &vdpasim->config + offset, len);
+ memcpy(buf, (u8 *)&vdpasim->config + offset, len);
}
static void vdpasim_set_config(struct vdpa_device *vdpa, unsigned int offset,
--
2.25.1
On 2020/4/6 ??10:45, Dan Carpenter wrote:> The "read" and "write" variables need to be signed for the error > handling to work. > > Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator") > Signed-off-by: Dan Carpenter <dan.carpenter at oracle.com> > --- > drivers/vdpa/vdpa_sim/vdpa_sim.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c > index 6e8a0cf2fdeb..b3c800653056 100644 > --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c > +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c > @@ -132,7 +132,8 @@ static void vdpasim_work(struct work_struct *work) > vdpasim, work); > struct vdpasim_virtqueue *txq = &vdpasim->vqs[1]; > struct vdpasim_virtqueue *rxq = &vdpasim->vqs[0]; > - size_t read, write, total_write; > + ssize_t read, write; > + size_t total_write; > int err; > int pkts = 0; >Acked-by: Jason Wang <jasowang at redhat.com>
Jason Wang
2020-Apr-09 02:06 UTC
[PATCH 2/2] vdpa: Fix pointer math bug in vdpasim_get_config()
On 2020/4/6 ??10:45, Dan Carpenter wrote:> If "offset" is non-zero then we end up copying from beyond the end of > the config because of pointer math. We can fix this by casting the > struct to a u8 pointer. > > Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator") > Signed-off-by: Dan Carpenter <dan.carpenter at oracle.com> > --- > Is it really worth letting people specify the offset? > > drivers/vdpa/vdpa_sim/vdpa_sim.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c > index b3c800653056..e03c25765513 100644 > --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c > +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c > @@ -509,7 +509,7 @@ static void vdpasim_get_config(struct vdpa_device *vdpa, unsigned int offset, > struct vdpasim *vdpasim = vdpa_to_sim(vdpa); > > if (offset + len < sizeof(struct virtio_net_config)) > - memcpy(buf, &vdpasim->config + offset, len); > + memcpy(buf, (u8 *)&vdpasim->config + offset, len); > } > > static void vdpasim_set_config(struct vdpa_device *vdpa, unsigned int offset,Acked-by: Jason Wang <jasowang at redhat.com>