Rick Wagner wrote: > I have a P4 system, which >>appears<< to be HT capable. However, the BIOS > (newest from vendor) does not give the option of turning HT on. Without > support from the BIOS, can HT be forced? Or am I mis-reading the info below, > and my CPU really is not HT capable. > > BTW, I tested with 2.6.10-1.741_FC3smp, and tried "acpismp=force" (as > described below). Thiago Guzella wrote: > Well, /proc/cpuinfo reports the HT feature flag as active, so this is > a HT capable cpu... Wouldn't it be nice? Some of Intel's documentation suggests this, too. Unfortunately, Intel seems to consider that "hyperthreading" and "actually having two logical CPUs" are different concepts. References: http://www.intel.com/support/processors/sb/cs-009861.htm and linked PDF (500K): on page 17, it says: > This field does not indicate that Hyper-Threading Technology has been > enabled for this specific processor. To determine if Hyper-Threading > Technology is supported, check the value returned in EBX[23:16] after > executing CPUID with EAX=1. If EBX[23:16] contains a value >1, then > the processor supports Hyper-Threading Technology. The value in EBX[23:16] is a count of the number of siblings. Unfortunately, I can't see an easy way of getting at this without a C (or assembler) program. This is right inside the processor: even a changed BIOS couldn't change that. [1] But http://indigo.intel.com/compare_cpu/default.aspx?familyID=1 lists the 2.66 GHz Pentium 4 as not having hyperthreading. This probably wasn't what you wanted to hear. Sorry. James. [1] Pointless speculation: *if* this just means that there is a working but disabled second sibling logical processor in there, then conceivably a "fixed" microcode could re-enable the processor. But you'd need *very* good links with Intel to get such a thing. -- James Wilkinson | You will stop at nothing to reach your objective, Exeter Devon UK | but only because your brakes are defective. E-mail address: james | @westexe.demon.co.uk |
