On Fri, 2002-06-14 at 14:53, Mikael Pettersson wrote:
> Unless my memory is failing me, I believe the simplest approach
> is to (1) don't set CONFIG_X86_TSC, and (2) pass "notsc" as a
> kernel boot parameter.
Correct, and this patch basically does both of the above.
The problem is that CONFIG_X86_TSC is enabled on PPro and above cpus.
The machines which are having this problem are multi-node P3 or P4
systems. Each cpu has a working TSC, its just that because they are not
synced they should not be used.
So the patch adds a CONFIG_DISABLE_TSC which is then checked where
earlier just CONFIG_X86_TSC was used. Additionally, if
CONFIG_DISABLE_TSC is set, the flag set by "notsc" is also set.
The usage of CONFIG_X86_TSC took me a bit to get my head around
initially, so your clarification is helpful.
Thanks
-john
-
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/
This archive was generated by hypermail 2b29 : Sat Jun 15 2002 - 22:00:32 EST