Bug#631945: [Bugme-new] [Bug 39372] New: Problems with HFSC Scheduler
- To: eric.dumazet@gmail.com
- Cc: kaber@trash.net, soltys@ziu.info, akpm@linux-foundation.org, netdev@vger.kernel.org, bugme-daemon@bugzilla.kernel.org, hadi@cyberus.ca, lucas.bocchi@gmail.com, 631945@bugs.debian.org, 00bormoj@gmail.com, fdelawarde@wirelessmundi.com
- Subject: Bug#631945: [Bugme-new] [Bug 39372] New: Problems with HFSC Scheduler
- From: Michal Pokrywka <wolfmoon@o2.pl>
- Date: Thu, 04 Aug 2011 15:55:15 +0200
- Message-id: <[🔎] 344feff5.16ac6e1d.4e3aa4c3.49362@o2.pl>
- Reply-to: Michal Pokrywka <wolfmoon@o2.pl>, 631945@bugs.debian.org
From: Michal Soltys <soltys@ziu.info>
Date: Fri, 29 Jul 2011 14:29:10 +0200
>> On 11-07-22 11:58, Michal Pokrywka wrote:
>>
>> commit b262a5da755cc6ed0cb4fba230cd9bf4037e1096 sch_sfq: fix peek()
>> implementation
>>
>> and
>>
>> commit 9df49f2bfe862573911a080c75a6d81113c5c81d sch_sfq: avoid giving
>> spurious NET_XMIT_CN signals
>>
>> Reverting these patches makes HFSC work again.
>>
>
> This one (upstream 8efa885406359af300d46910642b50ca82c0fe47) seems to be
> the culprit (does reverting only that one cures the problem ?)
Yes, indeed. Commit b262a5da755cc6ed0cb4fba230cd9bf4037e1096 is irrelevant.
Patch "sch_sfq: fix sfq_enqueue()" (provided by Eric Dumazet) applied to 2.6.39.1 fixes the problem, and hfsc+sfq works as expected.
Reply to: