On Thursday 27 August 2009 11:05:26 Dominik Brodowski wrote: > On Thu, Aug 27, 2009 at 10:53:24AM -0400, Jarod Wilson wrote: > > On Thursday 27 August 2009 10:32:46 Andi Kleen wrote: > > > Jarod Wilson wrote: > > > > On Wednesday 26 August 2009 13:19:37 Prarit Bhargava wrote: > > > >> Updated patch to only fail when hyperthreading is enabled. > > > > > > > > Very nice, I rather like this version. > > > > > > It still seems very drastic, the big hammer. > > > > Better than hard-locking the box, no? > > > > > You should at least check if mces are enabled. > > > > My vague recollection is that the box locks up regardless, but I could > > be misremembering. > > But why does _this_ box lock up if this erratum hasn't been observed in the > wild elsewhere It takes a fairly specific workload to trigger, see Prarit's latest mail in the thread. > -- at least that's what the spec update says, IIRC? The spec update was put out before it had been observed in the wild. Now its been observed in the wild (well, both Red Hat and Intel labs). -- Jarod Wilson jarod@xxxxxxxxxx -- To unsubscribe from this list: send the line "unsubscribe cpufreq" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
