Virtualization - Feb 2021 - [PATCH v2] vhost vdpa: fix the missing irq_bypass_unregister_producer() invocation

When qemu with vhost-vdpa netdevice is run for the first time, it works
well.
But after the VM is powered off, the next qemu run causes kernel panic due
to a
NULL pointer dereference in irq_bypass_register_producer().

When the VM is powered off, vhost_vdpa_clean_irq() misses on calling
irq_bypass_unregister_producer() for irq 0 because of the existing check.

This leaves stale producer nodes, which are reset in
vhost_vring_call_reset()
when vhost_dev_init() is invoked during the second qemu run.

As the node member of struct irq_bypass_producer is also initialized
to zero, traversal on the producers list causes crash due to NULL pointer
dereference.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=211711
Signed-off-by: Gautam Dawar <gdawar at xilinx.com>

-----
changelog:
v1->v2:
 - Addressed Jason's comment to remove the irq check and use
   vhost_vdpa_unsetup_vq_irq() to avoid local variable vq
-----


diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c
index 62a9bb0efc55..e00573b87aba 100644
--- a/drivers/vhost/vdpa.c
+++ b/drivers/vhost/vdpa.c
@@ -844,14 +844,10 @@ static int vhost_vdpa_open(struct inode *inode,
struct file *filep)

 static void vhost_vdpa_clean_irq(struct vhost_vdpa *v)
 {
-       struct vhost_virtqueue *vq;
        int i;

-       for (i = 0; i < v->nvqs; i++) {
-               vq = &v->vqs[i];
-               if (vq->call_ctx.producer.irq)
-
irq_bypass_unregister_producer(&vq->call_ctx.producer);
-       }
+       for (i = 0; i < v->nvqs; i++)
+               vhost_vdpa_unsetup_vq_irq(v, i);
 }

 static int vhost_vdpa_release(struct inode *inode, struct file *filep)
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.linuxfoundation.org/pipermail/virtualization/attachments/20210223/0d65f468/attachment-0001.html>

On 2021/2/23 6:00 ??, Gautam Dawar wrote:
> When qemu with vhost-vdpa netdevice is run for the first time, it 
> works well.
> But after the VM is powered off, the next qemu run causes kernel panic 
> due to a
> NULL pointer dereference in irq_bypass_register_producer().
>
> When the VM is powered off, vhost_vdpa_clean_irq() misses on calling
> irq_bypass_unregister_producer() for irq 0 because of the existing check.
>
> This leaves stale producer nodes, which are reset in 
> vhost_vring_call_reset()
> when vhost_dev_init() is invoked during the second qemu run.
>
> As the node member of struct irq_bypass_producer is also initialized
> to zero, traversal on the producers list causes crash due to NULL pointer
> dereference.
>
> Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=211711 
> <https://bugzilla.kernel.org/show_bug.cgi?id=211711>
> Signed-off-by: Gautam Dawar <gdawar at xilinx.com <mailto:gdawar at
xilinx.com>>
>
> -----
> changelog:
> v1->v2:
> ?- Addressed Jason's comment to remove the irq check and use
> ? ?vhost_vdpa_unsetup_vq_irq() to avoid local variable vq
> -----
>
Acked-by: Jason Wang <jasowang at redhat.com>

>
> diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c
> index 62a9bb0efc55..e00573b87aba 100644
> --- a/drivers/vhost/vdpa.c
> +++ b/drivers/vhost/vdpa.c
> @@ -844,14 +844,10 @@ static int vhost_vdpa_open(struct inode *inode, 
> struct file *filep)
>
> ?static void vhost_vdpa_clean_irq(struct vhost_vdpa *v)
> ?{
> - ? ? ? struct vhost_virtqueue *vq;
> ? ? ? ? int i;
>
> - ? ? ? for (i = 0; i < v->nvqs; i++) {
> - ? ? ? ? ? ? ? vq = &v->vqs[i];
> - ? ? ? ? ? ? ? if (vq->call_ctx.producer.irq)
> - irq_bypass_unregister_producer(&vq->call_ctx.producer);
> - ? ? ? }
> + ? ? ? for (i = 0; i < v->nvqs; i++)
> + ? ? ? ? ? ? ? vhost_vdpa_unsetup_vq_irq(v, i);
> ?}
>
> ?static int vhost_vdpa_release(struct inode *inode, struct file *filep)
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.linuxfoundation.org/pipermail/virtualization/attachments/20210223/074e9a68/attachment-0001.html>

On Tue, Feb 23, 2021 at 03:30:45PM +0530, Gautam Dawar
wrote:
> When qemu with vhost-vdpa netdevice is run for the first time, it works
well.
> But after the VM is powered off, the next qemu run causes kernel panic due
to a
> NULL pointer dereference in irq_bypass_register_producer().
> 
> When the VM is powered off, vhost_vdpa_clean_irq() misses on calling
> irq_bypass_unregister_producer() for irq 0 because of the existing check.
> 
> This leaves stale producer nodes, which are reset in
vhost_vring_call_reset()
> when vhost_dev_init() is invoked during the second qemu run.
> 
> As the node member of struct irq_bypass_producer is also initialized
> to zero, traversal on the producers list causes crash due to NULL pointer
> dereference.
> 
> Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=211711
> Signed-off-by: Gautam Dawar <gdawar at xilinx.com>
Could you please repost without HTML? Otherwise git am fails on this patch:

error: corrupt patch at line 6
error: could not build fake ancestor
Patch failed at 0001 vhost vdpa: fix the missing
irq_bypass_unregister_producer() invocation

Also, which commit introduces the bug? A Fixes: tag would be helpful.

Thanks!
> -----
> changelog:
> v1->v2:
> ?- Addressed Jason's comment to remove the irq check and use
> ? ?vhost_vdpa_unsetup_vq_irq() to avoid local variable vq
> -----
> 
> 
> diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c
> index 62a9bb0efc55..e00573b87aba 100644
> --- a/drivers/vhost/vdpa.c
> +++ b/drivers/vhost/vdpa.c
> @@ -844,14 +844,10 @@ static int vhost_vdpa_open(struct inode *inode,
struct
> file *filep)
> 
> ?static void vhost_vdpa_clean_irq(struct vhost_vdpa *v)
> ?{
> - ? ? ? struct vhost_virtqueue *vq;
> ? ? ? ? int i;
> 
> - ? ? ? for (i = 0; i < v->nvqs; i++) {
> - ? ? ? ? ? ? ? vq = &v->vqs[i];
> - ? ? ? ? ? ? ? if (vq->call_ctx.producer.irq)
> - ? ? ? ? ? ? ? ? ? ? ?
irq_bypass_unregister_producer(&vq->call_ctx.producer);
> - ? ? ? }
> + ? ? ? for (i = 0; i < v->nvqs; i++)
> + ? ? ? ? ? ? ? vhost_vdpa_unsetup_vq_irq(v, i);
> ?}
> 
> ?static int vhost_vdpa_release(struct inode *inode, struct file *filep)