netdev - Re: [PATCH 4/4] net_sched: gred: actually perform idling in WRED mode

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <50523632.60703@ll.mit.edu>
Date:	Thu, 13 Sep 2012 15:38:26 -0400
From:	"Ward, David - 0663 - MITLL" <david.ward@...mit.edu>
To:	Jamal Hadi Salim <jhs@...atatu.com>
CC:	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	Bruce Osler <brosler@...co.com>,
	Cyril Chemparathy <cyril@...com>
Subject: Re: [PATCH 4/4] net_sched: gred: actually perform idling in WRED
 mode

On 13/09/12 14:08, Jamal Hadi Salim wrote:
> On 12-09-13 11:22 AM, David Ward wrote:
>> gred_dequeue() and gred_drop() do not seem to get called when the
>> queue is empty, meaning that we never start idling while in WRED
>> mode. And since qidlestart is not stored by gred_store_wred_set(),
>> we would never stop idling while in WRED mode if we ever started.
>> This messes up the average queue size calculation that influences
>> packet marking/dropping behavior.
>>
>> Now, we start WRED mode idling as we are removing the last packet
>> from the queue. Also we now actually stop WRED mode idling when we
>> are enqueuing a packet.
>>
>> Cc: Bruce Osler <brosler@...co.com>
>> Signed-off-by: David Ward <david.ward@...mit.edu>
> This is one is not so obvious. Iam assuming you vetted it via some tests.
> In which case:
> Acked-by: Jamal Hadi Salim <jhs@...atatu.com>
>
> cheers,
> jamal
>

Before applying this patch, the average queue size (as seen with "tc -s 
qdisc") remained constant forever after I stopped sending any packets 
through the interface -- it didn't taper off as you would expect.  After 
the patch, the average queue size will now taper off if packets are not 
being sent.

David


Download attachment "smime.p7s" of type "application/pkcs7-signature" (4571 bytes)