Date: Sat, 1 May 2004 15:54:22 +1000 (EST) From: Bruce Evans <bde@zeta.org.au> To: Mike Silbersack <silby@silby.com> Cc: John Baldwin <jhb@FreeBSD.org> Subject: Re: cvs commit: src/sys/i386/isa clock.c Message-ID: <20040501154619.W19822@gamplex.bde.org> In-Reply-To: <20040430220507.B8726@odysseus.silby.com> References: <200404272003.i3RK3RFZ048001@repoman.freebsd.org> <20040430220507.B8726@odysseus.silby.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 30 Apr 2004, Mike Silbersack wrote: > On Tue, 27 Apr 2004, John Baldwin wrote: > > > jhb 2004/04/27 13:03:27 PDT > > > > FreeBSD src repository > > > > Modified files: > > sys/i386/isa clock.c > > Log: > > - The i8254 uses IRQ 0, not IRQ 8. Correct i8254_intsrc to reference the > > correct interrupt source. > > - Cache a pointer to the i8254_intsrc's pending method to avoid several > > pointer indirections in i8254_get_timecount(). > > Does this fix a bug with clock handling? I ask because my system's clock > has been running fast recently, and I haven't had time to investigate why > that is occuring... This fixes breakage of the i8254 timecounter last December by the new interrupt code. Counter overflow was not detected properly, so there was sometimes a false negative which caused the clock to jump by 1/HZ seconds. I think false positives were also possible, and the jump for fals negatives made the clock run fast. However, the i8254 timecounter is rarely used on new systems. It is now used mainly on SMP systems without acpi. Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040501154619.W19822>