From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tim Sander Subject: Re: [ANNOUNCE] 4.1.5-rt5 meant to reply to 4.4.1-rt5 Date: Fri, 12 Feb 2016 15:36:16 +0100 Message-ID: <7138155.u6nNkmxGuH@dabox> References: <20150816135630.GE7004@linutronix.de> <1614199.PesBpXyHut@dabox> <56BDA0EF.7090503@linutronix.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-rt-users , LKML , Thomas Gleixner , rostedt@goodmis.org, John Kacur To: Sebastian Andrzej Siewior Return-path: In-Reply-To: <56BDA0EF.7090503@linutronix.de> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-rt-users.vger.kernel.org Hi Sebastian As you got correctly i was talking about 4.4.1-rt5 and not 4.1 i replie= d to by=20 accident. Am Freitag, 12. Februar 2016, 10:07:59 schrieb Sebastian Andrzej Siewio= r: > On 02/12/2016 09:28 AM, Tim Sander wrote: > > Hi Sebastian >=20 > Hi Tim, >=20 > > Am Sonntag, 16. August 2015, 15:56:30 schrieb Sebastian Andrzej Sie= wior: > >> I'm pleased to announce the v4.1.5-rt5 patch set. > >=20 > > I have just tested it with a Altera SoC ARM v7. The latencies seem = to have > > gotten a little bit worse with each release. The first core has alw= ays > > been > > worse (presumably due to interrupt load) but now it dropped to 111=B5= s (rt5) > > from 76=B5s(rt3) and 54=B5s(rt2). >=20 > in -rt2 we had bug in migrate disable code which means each task was > running on CPU0. This got partly fixed in -rt3. In -rt3 the scheduler > could assign a task to CPU1 but the task should stay there for ever. > This little detail was fixed in -rt5. > This is one thing that comes to mind. > Lazy-preempt should have been fixed in -rt3, too. This should not giv= e > you higher latencies but higher throughput. >=20 > What about rt4? It is only the stable update so you should see here t= he > numbers from rt3. If that is true and your numbers are stable it shou= ld > be easy to run git bisect between rt4 and rt5. And looking at > https://git.kernel.org/rt/linux-rt-devel/h/v4.4.1-rt5 > the only non-cosmetic change in -rt5 that should affect you is the > migrate-disable fixup from Mike. Ok, each run takes a couple of hours so bisecting should take quite som= e time but i will give it a try. I started a test with 4.4.1-rt4, if the numbe= rs are=20 within the 70=B5s ballpark bisecting seems the way to go. If the number= s are=20 higher i suspect that stable update might have a play here. But we will= see. Best regards Tim