Re: RFC: Bogus setting of CONSTANT TSC ?

From: Andi Kleen
Date: Sun Mar 13 2011 - 13:06:43 EST


> This was initially introduced with commit 39b3a7 ([PATCH] i386/x86-64:
> Generalize X86_FEATURE_CONSTANT_TSC flag), though the changelog is
> utterly useless. Andi ??

Only Nehalem and beyond set the 8000_0007 flag, but Intel has had a constant
TSC since Yonah. That is what this logic implements.

However on Nehalem and beyond if the 8000_0007 is clear the BIOS wants
to tell us that it knows the system doesn't have a synchronized TSC.
This is currently still broken, but the timer watchdog usually catches
it later. It only affects a few relatively obscure systems.

-Andi
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/