Re: [PATCH 3/11] arch/x86/kernel: Correct NULL test

From: Ludovic FERRE
Date: Sat Feb 06 2010 - 17:07:25 EST


Hello Julia,

May be it would be worthy to take your changes one step further and to
report if the function is called with a null device pointer (which
indicates a problem in the calling code, right?).

Here's a patch moving the null pointer check to its own if block and
reporting the null pointer function call using printk based on
2.6.33-rc6 code.

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 23824fe..1a1d87c 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -114,11 +114,17 @@ static bool check_device(struct device *dev)
{
u16 devid;

- if (!dev || !dev->dma_mask)
+ /* Function is called with a null pointer. Flag this and return false */
+ if (!dev){
+ printk(KERN_INFO "amd_iommu check_device called with a
null pointer device\n");
+ return false;
+ }
+
+ if (!dev->dma_mask)
return false;

/* No device or no PCI device */
- if (!dev || dev->bus != &pci_bus_type)
+ if (dev->bus != &pci_bus_type)
return false;

devid = get_device_id(dev);

On Sat, Feb 6, 2010 at 9:42 AM, Julia Lawall <julia@xxxxxxx> wrote:
> From: Julia Lawall <julia@xxxxxxx>
>
> dev was tested just above, so drop the second test.
>
> A simplified version of the semantic match that finds this problem is as
> follows: (http://coccinelle.lip6.fr/)
>
> // <smpl>
> @r@
> expression *x;
> expression e;
> identifier l;
> @@
>
> if (x == NULL || ...) {
> Â Â... when forall
> Â Âreturn ...; }
> ... when != goto l;
> Â Âwhen != x = e
> Â Âwhen != &x
> *x == NULL
> // </smpl>
>
> Signed-off-by: Julia Lawall <julia@xxxxxxx>
>
> ---
> Âarch/x86/kernel/amd_iommu.c     |  Â2 +-
> Â1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
> index adb0ba0..2c4a501 100644
> --- a/arch/x86/kernel/amd_iommu.c
> +++ b/arch/x86/kernel/amd_iommu.c
> @@ -118,7 +118,7 @@ static bool check_device(struct device *dev)
> Â Â Â Â Â Â Â Âreturn false;
>
> Â Â Â Â/* No device or no PCI device */
> - Â Â Â if (!dev || dev->bus != &pci_bus_type)
> + Â Â Â if (dev->bus != &pci_bus_type)
> Â Â Â Â Â Â Â Âreturn false;
>
> Â Â Â Âdevid = get_device_id(dev);
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at Âhttp://vger.kernel.org/majordomo-info.html
> Please read the FAQ at Âhttp://www.tux.org/lkml/
>



--
Ludovic
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/