Date: Tue, 20 Jun 2006 20:15:39 +0300 From: Andriy Gapon <avg@icyb.net.ua> To: freebsd-current@freebsd.org Cc: Nate Lawson <nate@root.org> Subject: Re: Freeze due to performance_cx_lowest=LOW Message-ID: <44982D3B.9050409@icyb.net.ua> In-Reply-To: <4497F3DE.7030000@icyb.net.ua> References: <1148837064.00534930.1148826605@10.7.7.3> <200606011504.31635.jhb@freebsd.org> <44929864.4080207@icyb.net.ua> <200606161509.28998.jhb@freebsd.org> <4497F3DE.7030000@icyb.net.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
on 20/06/2006 16:10 Andriy Gapon said the following: > Well, I think MP systems are "meant" to use lapic timer; anyway, I do > not care much for them now :-) But I am exploring possibilities to avoid > using lapic timer (or to augment it) on UP systems. And, as you confirm, > RTC seems to be a good candidate for that. > BTW, an interesting link here: http://www.microsoft.com/whdc/system/CEC/mm-timer.mspx Some quotes: 8254 PIT ... this timer is only used in periodic mode to provide the periodic clock interrupt on uni-processor systems. [in Windows] RTC ... The RTC is used in periodic mode to provide the system profiling interrupt on uni-processor systems and the clock interrupt on multi-processor systems. [in Windows] Not sure about their terminology and how up-to-date this document is, but I think that they use 8254 for hardclock on non-APIC systems and RTC on APIC systems in Windows XP. -- Andriy Gapon
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44982D3B.9050409>