Re: [Xen-devel] [PATCH 1/9] xen/pciback: Do not dereference psdev during printk when it is NULL.

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 29.09.11 at 21:52, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx> wrote:
> . instead use printk(.. facility.
> 
> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
> ---
>  drivers/xen/xen-pciback/pci_stub.c |    8 +++++---
>  1 files changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/xen/xen-pciback/pci_stub.c 
> b/drivers/xen/xen-pciback/pci_stub.c
> index aec214a..32d6891 100644
> --- a/drivers/xen/xen-pciback/pci_stub.c
> +++ b/drivers/xen/xen-pciback/pci_stub.c
> @@ -514,9 +514,11 @@ static void kill_domain_by_device(struct pcistub_device 
> *psdev)
>       int err;
>       char nodename[PCI_NODENAME_MAX];
>  
> -     if (!psdev)
> -             dev_err(&psdev->dev->dev,
> -                     "device is NULL when do AER recovery/kill_domain\n");
> +     if (!psdev) {
> +             printk(KERN_ERR DRV_NAME
> +                     ":device is NULL when do AER recovery/kill_domain\n");
> +             return;
> +     }

This is bogus - all callers of this function already make sure psdev is
non-NULL, so imo the check should be removed or replaced with a
BUG_ON().

Jan

>       snprintf(nodename, PCI_NODENAME_MAX, "/local/domain/0/backend/pci/%d/0",
>               psdev->pdev->xdev->otherend_id);
>       nodename[strlen(nodename)] = '\0';





_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel