From 272870cf7bbcb8fb0204ec80e2ef0c6849eabd23 Mon Sep 17 00:00:00 2001 From: Poul-Henning Kamp Date: Mon, 17 Mar 2008 09:05:15 +0000 Subject: [PATCH] A cautionary XXX comment about seemingly bogus errata checks. --- sys/i386/cpufreq/p4tcc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/sys/i386/cpufreq/p4tcc.c b/sys/i386/cpufreq/p4tcc.c index 501aed581f4..4dc206a9e07 100644 --- a/sys/i386/cpufreq/p4tcc.c +++ b/sys/i386/cpufreq/p4tcc.c @@ -164,6 +164,14 @@ p4tcc_attach(device_t dev) */ sc->auto_mode = TRUE; + /* + * XXX: After a cursory glance at various Intel specification + * XXX: updates it seems like these tests for errata is bogus. + * XXX: As far as I can tell, the failure mode is benign, in + * XXX: that cpus with no errata will have their bottom two + * XXX: STPCLK# rates disabled, so rather than waste more time + * XXX: hunting down intel docs, just document it and punt. /phk + */ switch (cpu_id & 0xff) { case 0x22: case 0x24: