1 |
As someone else mentioned, my guess would be it's turned off on the board. If there's no bios setting, how about a jumper on the board? There has to be some way to turn it on/off. |
2 |
|
3 |
|
4 |
-----Original Message----- |
5 |
From: A. Khattri [mailto:ajai@××××.net] |
6 |
Sent: Sun 9/25/2005 1:36 AM |
7 |
To: gentoo-server@l.g.o |
8 |
Cc: |
9 |
Subject: Re: [gentoo-server] HT on P4? |
10 |
On Fri, 23 Sep 2005, Joao Patricio wrote: |
11 |
|
12 |
> Did you enable SMP when you compiled your kernel? |
13 |
|
14 |
Both SMP and SMT are enabled in this kernel: |
15 |
|
16 |
# grep SMP /usr/src/linux/.config |
17 |
|
18 |
# CONFIG_X86_BIGSMP is not set |
19 |
CONFIG_SMP=y |
20 |
CONFIG_X86_FIND_SMP_CONFIG=y |
21 |
CONFIG_X86_SMP=y |
22 |
|
23 |
# grep SMT /usr/src/linux/.config |
24 |
|
25 |
CONFIG_SCHED_SMT=y |
26 |
|
27 |
|
28 |
|
29 |
> |
30 |
> Em Sexta, 23 de Setembro de 2005 23:05, o A. Khattri escreveu: |
31 |
> > On Fri, 23 Sep 2005, kashani wrote: |
32 |
> > > A. Khattri wrote: |
33 |
> > > > I switched on SMP and HT when building my kernel but I see this among |
34 |
> > > > the boot messages: |
35 |
> > > > |
36 |
> > > > CPU: After generic identify, caps: bfebfbff 00000000 00000000 00000000 |
37 |
> > > > 0000441d 00000000 00000000 |
38 |
> > > > CPU: After vendor identify, caps: bfebfbff 00000000 00000000 00000000 |
39 |
> > > > 0000441d 00000000 00000000 |
40 |
> > > > monitor/mwait feature present. |
41 |
> > > > using mwait in idle threads. |
42 |
> > > > CPU: Trace cache: 12K uops, L1 D cache: 16K |
43 |
> > > > CPU: L2 cache: 1024K |
44 |
> > > > CPU: Hyper-Threading is disabled |
45 |
> > > |
46 |
> > > Is it possible it's disabled in the Bios? Some servers shipped that way |
47 |
> > > when they first came out IIRC. |
48 |
> > |
49 |
> > I didn't find any BIOS setting unfortunately... |
50 |
> > |
51 |
> > This is a Supermicro 5013C-T server with a P4 (P4SCE?) board in it. |
52 |
> > |
53 |
> > # cat /proc/cpuinfo |
54 |
> > processor : 0 |
55 |
> > vendor_id : GenuineIntel |
56 |
> > cpu family : 15 |
57 |
> > model : 4 |
58 |
> > model name : Intel(R) Pentium(R) 4 CPU 2.40GHz |
59 |
> > stepping : 1 |
60 |
> > cpu MHz : 2395.003 |
61 |
> > cache size : 1024 KB |
62 |
> > fdiv_bug : no |
63 |
> > hlt_bug : no |
64 |
> > f00f_bug : no |
65 |
> > coma_bug : no |
66 |
> > fpu : yes |
67 |
> > fpu_exception : yes |
68 |
> > cpuid level : 5 |
69 |
> > wp : yes |
70 |
> > flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca |
71 |
> > cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe pni monitor |
72 |
> > ds_cpl cid xtpr |
73 |
> > bogomips : 4718.59 |
74 |
> > |
75 |
> > |
76 |
> > -- |
77 |
> |
78 |
|
79 |
-- |
80 |
wheel of reincarnation |
81 |
|
82 |
[coined in a paper by T. H. Myer |
83 |
and I.E. Sutherland "On the Design of Display Processors", Comm. |
84 |
ACM, Vol. 11, no. 6, June 1968)] Term used to refer to a well-known |
85 |
effect whereby function in a computing system family is migrated |
86 |
out to special-purpose peripheral hardware for speed, then the |
87 |
peripheral evolves toward more computing power as it does its job, |
88 |
then somebody notices that it is inefficient to support two |
89 |
asymmetrical processors in the architecture and folds the function |
90 |
back into the main CPU, at which point the cycle begins again. |
91 |
|
92 |
Several iterations of this cycle have been observed in |
93 |
graphics-processor design, and at least one or two in |
94 |
communications and floating-point processors. Also known as `the |
95 |
Wheel of Life', `the Wheel of Samsara', and other variations of |
96 |
the basic Hindu/Buddhist theological idea. See also blitter, |
97 |
bit bang. |
98 |
|
99 |
-- |
100 |
gentoo-server@g.o mailing list |