Date: Thu, 10 Sep 2015 00:55:02 +0200 (CEST) From: Andreas Schwarz <freebsd.asc@strcmp.org> To: freebsd-arm@FreeBSD.org Subject: Re: DWC OTG TX path optimisation for 11-current Message-ID: <46e495e484.66704618@mail.schwarzes.net> In-Reply-To: <46d8b55830c.48a059ec@mail.schwarzes.net> References: <55A7D8CE.4020809@selasky.org> <CAHNYxxMp9jGDbV-5=-cE6daR-O3eN5pdvO1s-=QfX=A9XYqYmA@mail.gmail.com> <55B23276.8090703@selasky.org> <CAHNYxxNc9uB62hHEv1PM9PcsGgUs=zsvNgatqLD0p%2BiiDA3Aiw@mail.gmail.com> <55B73113.2020308@selasky.org> <CAFHCsPVaPZpqXLS7OApa=Xz5VLnLjVpV5dYV8Pn2uHh1Lcz7Tg@mail.gmail.com> <55B8AB76.7030603@selasky.org> <CAFHCsPUMaYEwJsaGUFuw9yZi_5bmraSBsOYpRWvSeuebpXBJUA@mail.gmail.com> <55B8B297.1010008@selasky.org> <CAFHCsPVGLs8j6LAV%2Bg4rP_ueTOd8pUOupYFGvmgC3XGcJC720Q@mail.gmail.com> <20150729154516.GH78154@funkthat.com> <55B8F5EC.2050908@selasky.org> <46ad096c958.1a82a175@mail.schwarzes.net> <55B9C3E2.5040501@selasky.org> <46ae815c7c3.447237c8@mail.schwarzes.net> <46aece00b53.3c1cdc1f@mail.schwarzes.net> <55BB2A5F.9000502@selasky.org> <46baa16c4ce.6efd29ef@mail.schwarzes.net> <55CF31A1.5080205@selasky.org> <46ce372c895.20050775@mail.schwarzes.net> <46d0a4441bb.41f6f91d@mail.schwarzes.net> <55DD5C0A.2050401@selasky.org> <CAFHCsPVTpnd2RB4d0NvLRf9WniSkBPajJYABoPOuS4NKo%2BZ7PA@mail.gmail.com> <46d8b55830c.48a059ec@mail.schwarzes.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 01.09.15, Andreas Schwarz wrote: > On 28.08.15, Svatopluk Kraus wrote: > >> My subjective observation is that the slow system response can be >> caused by slow wake up from idle state. I have tried to change >> scheduler from ULE to BSD one with no MII warning and everything is >> okay result. And buildword was about one hour faster. Note that the >> trigger is very sensitive, so it can be just because system timing and >> many other things are different with different scheduler. > > I've tried the 4BSD scheduler too, amazing, faster and surely prevents > to run into the known problem. I've just started a buildworld loop, will > try to reach 10 rounds. I just want to inform you that I've copmpleted my ten rounds of building (with -j4) and cleaning the world in a loop without any problems (and INVARIANTS switched off) at my RPI2 (see script below). It take ~18 hours for a "round". It's not a fix but a real workaround to prevent this annoying interrupt problems with the sd driver. Seems, that the more "stolid" 4BSD scheduler will not drive the system in situations which trigger the problem. root@pizelot:~ # uname -a FreeBSD pizelot.schwarzes.net 11.0-CURRENT FreeBSD 11.0-CURRENT #0 r287315: Mon Aug 31 12:03:20 CEST 2015 root@pizelot.schwarzes.net:/usr/obj/usr/src/sys/RPI2-B-ASC arm #!/bin/sh cd /usr/src while true do make cleandir make -j4 kernel-toolchain make -j4 buildkernel make -j4 buildworld date >> loop_build.log done -asc
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?46e495e484.66704618>