> I disagree with Alan's recommendation.
So do I.
> The real problem is that the kernel confuses a CPU-level property
> (do the CPUs have TSCs?) with a system-level property (are the
> TSCs present and in sync?). CONFIG_X86_TSC really describes the
> latter property, for the former we have the cpu_has_tsc() macro.
Well, CONFIG_X86_TSC simply asserts we have TSCs present and in sync and
cpu_has_tsc is a run-time check for the same. The X86_FEATURE_TSC bit
shouldn't be set (and e.g. "notsc" takes care of this) unless TSCs work
correctly as it's both used internally and exported to the userland. For
low-level fiddling with TSCs one can use cpuid either directly or with the
cpuid driver.
-- + Maciej W. Rozycki, Technical University of Gdansk, Poland + +--------------------------------------------------------------+ + e-mail: macro@ds2.pg.gda.pl, PGP key available +- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/