Date: Sat, 01 Oct 2011 14:20:43 +0300 From: Alexander Motin <mav@FreeBSD.org> To: Adrian Chadd <adrian@freebsd.org> Cc: "Jayachandran C." <jchandra@freebsd.org>, svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r225892 - head/sys/mips/mips Message-ID: <4E86F78B.7050707@FreeBSD.org> In-Reply-To: <CAJ-Vmon32cVEVvC=3WJVmDkCUdyLWyec3sqU-ifzspVSPxedfg@mail.gmail.com> References: <201110010556.p915uQH6003016@svn.freebsd.org> <CA%2B7sy7BiRvTB79H9=y%2BS4jQ=%2BboW1bcDJn%2BBULMmJU9KLLVJ5A@mail.gmail.com> <CAJ-VmokAsDpjJLt%2BVJ2gDGX%2BiMAwZvL2TPaaAD_LRm-Yyquxig@mail.gmail.com> <CA%2B7sy7D6h5a08Q6yNfX6xSqwabDLzE5GLu5aV3fCMYQKn_4AoQ@mail.gmail.com> <CAJ-Vmon32cVEVvC=3WJVmDkCUdyLWyec3sqU-ifzspVSPxedfg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Adrian Chadd wrote: > On 1 October 2011 15:21, Jayachandran C. <jchandra@freebsd.org> wrote: > >> I would like to understand this, reverting r216862 would take out the >> critical_enter() and cpu_idleclock() which would keep the timer >> interrupts coming is as usual, this should not affect the event timer. > > I haven't verified that mav's timer stuff does the correct thing by > configuring the clock timer to occur every 1000hz in this instance. I > kinda hope so. When setting kern.eventtimer.idletick=1, you were effectively disabling cpu_idleclock()/cpu_activeclock() and you've properly received about 1127 timer interrupts per second (mix of 1000 of hz and 127 of stathz). Also when CPU is busy or has high interrupt/context switch rate, that happen automatically, so it definitely works. If you are talking about programming hardware timer for periodic 1000Hz rate, it doesn't happen automatically, because I doubt benefits it gives worth strict aliasing of hardclock() and statclock() it causes. If you like, you can do it manually via kern.eventtimer.periodic=1. -- Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4E86F78B.7050707>