Avi Kivity wrote:
> On 01/13/2010 11:32 AM, Andrew Morton wrote:
>> On Wed, 13 Jan 2010 11:22:39 +0200 Avi Kivity<avi@xxxxxxxxxx> wrote:
>>
>>
>>> On 01/13/2010 12:11 AM, akpm@xxxxxxxxxxxxxxxxxxxx wrote:
>>>
>>>> Subject: kvm/ia64: dereference of NULL pointer in set_pal_result()
>>>> From: Roel Kluin<roel.kluin@xxxxxxxxx>
>>>>
>>>> Do not dereference a NULL pointer
>>>>
>>>> diff -puN
>>>> arch/ia64/kvm/kvm_fw.c~kvm-ia64-dereference-of-null-pointer-in-set_pal_result
>>>> arch/ia64/kvm/kvm_fw.c ---
>>>> a/arch/ia64/kvm/kvm_fw.c~kvm-ia64-dereference-of-null-pointer-in-set_pal_result
>>>> +++ a/arch/ia64/kvm/kvm_fw.c @@ -75,9 +75,11 @@ static void
>>>> set_pal_result(struct kvm_vc struct exit_ctl_data *p;
>>>>
>>>> p = kvm_get_exit_data(vcpu);
>>>> - if (p&& p->exit_reason == EXIT_REASON_PAL_CALL) { + if (!p)
>>>> + return;
>>>> + if (p->exit_reason == EXIT_REASON_PAL_CALL) {
>>>> p->u.pal_data.ret = result;
>>>> - return ;
>>>> + return;
>>>> }
>>>> INIT_PAL_STATUS_UNIMPLEMENTED(p->u.pal_data.ret); }
>>>>
>>>>
>>>
>>> kvm_get_exit_data() cannot return a NULL pointer.
>>>
>> In that case set_pal_result() doesn't need to test for that.
>>
>> Roel looks for code along the lines of
>>
>> if (p)
>> ...
>>
>> *p;
>>
>
> I see. I think it's worthwhile to look deeper rather than converting
> things mechanically. After all, the code may have actually worked
> before the patch.
Originally, we want to dynamically allocate the exit_data, so needs the check. But now, it is allocated statically, so the check is unnecessary.
Xiantao
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
