Date: Mon, 11 Jan 1999 20:31:37 -0800 From: Mike Smith <mike@smith.net.au> To: Peter Jeremy <peter.jeremy@auss2.alcatel.com.au> Cc: steve@Watt.COM, freebsd-stable@FreeBSD.ORG Subject: Re: IRQ entropy causes panics? Message-ID: <199901120431.UAA00469@dingo.cdrom.com> In-Reply-To: Your message of "Tue, 12 Jan 1999 08:32:38 %2B1100." <99Jan12.083159est.40333@border.alcanet.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
> On Sat 9th January, Steve Watt <steve@Watt.COM> wrote: > >Is there a microtime() call already made at each interrupt entry? > No. > > > Something on > >the order of 5uS on a faster ISA system I tested on > > On a non-SMP P5 or later system RDTSC is a cheap and accurate > alternative. [The problem on SMP systems is that the TSCs aren't > synchronised so you need to know which CPU you are on to use it]. > > On 3.x, both microtime() and nanotime() will use RDTSC (and not > perform any ISA bus cycles) if a working TSC is found, SMP is not > defined and either it isn't an APM BIOS, or APM isn't compiled into > the kernel. The actual clock being used, together with the lowest > level overhead of calling it, is reported by default during boot, eg > 'Timecounter "i8254" frequency 1193182 Hz cost 2376 ns' Note that the TSC can't be relied upon, even if no APM BIOS was found. The default timecounter code in current can detect when the assumptions it's making about the progression of time are faulty, but Poul didn't want to make the compensation for this (which it can also make) either the default or automatic. Talk to him about that. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199901120431.UAA00469>