[ovs-dev] [PATCH] sFlow: clear the padding

[Francesco Fusco]

Neil,
did you have a chance to have a look at the patch?

Best,
Francesco

On 12/19/2013 06:28 PM, Ben Pfaff wrote:
> Hi Francesco.  Thanks for the patch.
>
> The sFlow code comes almost unmodified from InMon.  I'm adding Neil
> McKee from InMon to the thread to review this change.  Neil, can you
> take a look?  (And that reminds me, I need to review your patch too.
> I'll do that now.)
>
> Thanks,
>
> Ben.
>
> On Thu, Dec 19, 2013 at 06:16:24PM +0100, Francesco Fusco wrote:
>> putString pads the string to the 4-byte boundary without
>> clearing the "padded" memory. This patch simply set the
>> padding to zero.
>>
>> Signed-off-by: Francesco Fusco <ffusco at redhat.com>
>> ---
>>   lib/sflow_receiver.c | 4 ++++
>>   1 file changed, 4 insertions(+)
>>
>> diff --git a/lib/sflow_receiver.c b/lib/sflow_receiver.c
>> index 3e5a67a..e6fc9a7 100644
>> --- a/lib/sflow_receiver.c
>> +++ b/lib/sflow_receiver.c
>> @@ -198,6 +198,10 @@ inline static void putString(SFLReceiver *receiver, SFLString *s)
>>       putNet32(receiver, s->len);
>>       memcpy(receiver->sampleCollector.datap, s->str, s->len);
>>       receiver->sampleCollector.datap += (s->len + 3) / 4; /* pad to 4-byte boundary */
>> +    if ((s->len % 4) != 0){
>> +        u_int8_t padding = 4 - (s->len % 4);
>> +        memset(((u_int8_t*)receiver->sampleCollector.datap)-padding, 0, padding);
>> +    }
>>   }
>>
>>   inline static u_int32_t stringEncodingLength(SFLString *s) {