Re: [Lksctp-developers] Incorrect use of abs() in sctp_transport_update_rto()

[Schoch C. <e03...@st...>]

Hi,

some months ago I had the same problem. You can have a look at the  
following discussion:
http://sourceforge.net/mailarchive/message.php?msg_id=25888601
>
>> Hi,
>>
>> it seems to me that abs() is used incorrectly while calculating
>> tp->rttvar in sctp_transport_update_rto().  Its parameter is unsigned,
>> meaning it will never be negative.  This quickly leads to RTO taking
>> on extremely high values.  I suggest changing the expression from:
>>
>>   abs(tp->srtt - rtt)
>>
>> to:
>>
>>   abs((long) tp->srtt - rtt)
>>

This patch should fix this problem:

-----------------------------------------------------------------------

--- a/net/sctp/transport.c	2010-10-05 06:04:17.000000000 +0200
+++ b/net/sctp/transport.c	2010-10-05 06:09:09.000000000 +0200
@@ -334,7 +334,7 @@
		 * 1/8, rto_alpha would be expressed as 3.
		 */
		tp->rttvar = tp->rttvar - (tp->rttvar >> sctp_rto_beta)
-			+ ((abs(tp->srtt - rtt)) >> sctp_rto_beta);
+			+ (((__u32)abs((long)tp->srtt - (long)rtt)) >> sctp_rto_beta);
		tp->srtt = tp->srtt - (tp->srtt >> sctp_rto_alpha)
			+ (rtt >> sctp_rto_alpha);
	} else {

---------------------------------------------



>> or similar.  This happens in Linux because abs() is a macro that
>> behaves differently from the libc function.  In libc, the implicit
>> cast to int ensures this problem does not occur.
>
> abs() handles unsigned and signed longs, ints, shorts and chars.  For all
> input types abs() returns a signed long.
>

It does not. As you could see in the link above, I had the same  
behavior. There is a difference by using abs() with signed and  
unsigned values. In detail, the problem starts when rtt is larger than  
srtt and the result is a very large rttvar which pushes the RTO to  
maximum.

I could not figure out where the problem exactly was situated, but  
after upgrading my system to Ubuntu 10.10 this problem has gone. I  
assumed it could be related to gcc or something like this.

Regards,
Christian

>
>
> ------------------------------------------------------------------------------
> Colocation vs. Managed Hosting
> A question and answer guide to determining the best fit
> for your organization - today and in the future.
> http://p.sf.net/sfu/internap-sfd2d
> _______________________________________________
> Lksctp-developers mailing list
> Lks...@li...
> https://lists.sourceforge.net/lists/listinfo/lksctp-developers
>