Prev: fatal trap 12 page fault while in kernel mode
Next: 8.0 sysinstall crashes
From: Torfinn Ingolfsen on 12 Jun 2010 05:57
A.Reader wrote:
> I need to get a higher-performance tame server for development,
> and it'd be quite nice to get one that uses the Xeon chip.
>
> But it looks like that'd be a kludge -- from what I can find on
> the site it seems to rely on the AMD64 build not noticing that
> the Xeon isn't an AMD chip.

Huh?
AMD64, X86-64, its all the same:
http://en.wikipedia.org/wiki/Amd64#AMD64
http://en.wikipedia.org/wiki/X86-64
--
Torfinn Ingolfsen,
Norway
From: Lowell Gilbert on 12 Jun 2010 11:58
A.Reader <anonymously(a)example.com> writes:

> On Sat, 12 Jun 2010 11:57:13 +0200,
> Torfinn Ingolfsen <tingo(a)start.no> wrote:
>
>>A.Reader wrote:
>>> I need to get a higher-performance tame server for development,
>>> and it'd be quite nice to get one that uses the Xeon chip.
>>>
>>> But it looks like that'd be a kludge -- from what I can find on
>>> the site it seems to rely on the AMD64 build not noticing that
>>> the Xeon isn't an AMD chip.
>>
>>Huh?
>>AMD64, X86-64, its all the same:
>>http://en.wikipedia.org/wiki/Amd64#AMD64
>>http://en.wikipedia.org/wiki/X86-64
>
> Perhaps I'm just not reading it correctly, but I'd expect to see
> some statement like "operating systems can't tell the difference
> between the two chips", or "the two chips are interchangeable as
> far as the operating system is concerned". What I seem to be
> seeing is that they're "alike" in some ways, but not identical.
>
> Am I just missing it?

What you're missing is a little more subtle. amd64 is an architecture,
not a chip. Xeon and Athlon64 are both examples of chips that use the
architecture.

--
Lowell Gilbert, embedded/networking software engineer
http://be-well.ilk.org/~lowell/
From: Andrew Reilly on 14 Jun 2010 07:03
On Sat, 12 Jun 2010 05:24:24 -0400, A.Reader wrote:

> Is there anyone with solid experience running fBSD on the Xeon, who
> could give me a rundown on what/how bad the issues are?

I've run 7, 8 and 9 on both AMD64 (Athlon flavour, not Opteron) and Xeon
(both NetBurst and Core2 flavour). Works as advertised, without
incident. At least, without incident related to the AMD-ness or Intel-
ness of the processor, anyway. (I've used intel boards with unhappy
Broadcom ethernet chips and AMD boards with unhappy NVidia chips: it's
not hard to find low-level discomfort...)

I haven't ever felt the need to do virtualization on these systems:
apparently that's one area where the two diverge.

Just one data-point for you...

--
Andrew
From: Torfinn Ingolfsen on 14 Jun 2010 11:45
On 06/12/2010 13:34, A.Reader wrote:
> On Sat, 12 Jun 2010 11:57:13 +0200,
> Torfinn Ingolfsen<tingo(a)start.no> wrote:
>
>> A.Reader wrote:
>>> I need to get a higher-performance tame server for development,
>>> and it'd be quite nice to get one that uses the Xeon chip.
>>>
>>> But it looks like that'd be a kludge -- from what I can find on
>>> the site it seems to rely on the AMD64 build not noticing that
>>> the Xeon isn't an AMD chip.
>>
>> Huh?
>> AMD64, X86-64, its all the same:
>> http://en.wikipedia.org/wiki/Amd64#AMD64
>> http://en.wikipedia.org/wiki/X86-64
>
> Perhaps I'm just not reading it correctly, but I'd expect to see
> some statement like "operating systems can't tell the difference
> between the two chips", or "the two chips are interchangeable as
> far as the operating system is concerned". What I seem to be
> seeing is that they're "alike" in some ways, but not identical.
>
> Am I just missing it?

Yes, probably.
Here is a datapoint:
root(a)kg-v7# uname -a
FreeBSD kg-v7.kg4.no 8.0-STABLE FreeBSD 8.0-STABLE #2: Sat May 8
13:50:00 CEST 2010 root(a)kg-v7.kg4.no:/usr/obj/usr/src/sys/GENERIC amd64
root(a)kg-v7# sysctl hw.machine
hw.machine: amd64
root(a)kg-v7# sysctl hw.model
hw.model: Intel(R) Core(TM) i3 CPU 540 @ 3.07GHz

Output from /var/log/messages
Mar 8 02:01:32 kg-v7 kernel: FreeBSD 8.0-RELEASE #0: Sat Nov 21
15:02:08 UTC 2009
Mar 8 02:01:32 kg-v7 kernel:
root(a)mason.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC
Mar 8 02:01:32 kg-v7 kernel: Timecounter "i8254" frequency 1193182 Hz
quality 0
Mar 8 02:01:32 kg-v7 kernel: CPU: Intel(R) Core(TM) i3 CPU 540
@ 3.07GHz (3066.68-MHz K8-class CPU)
Mar 8 02:01:32 kg-v7 kernel: Origin = "GenuineIntel" Id = 0x20652
Stepping = 2
Mar 8 02:01:32 kg-v7 kernel:
Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
Mar 8 02:01:32 kg-v7 kernel:
Features2=0x98e3bd<SSE3,DTES64,MON,DS_CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,SSE4.1,SSE4.2,POPCNT>
Mar 8 02:01:32 kg-v7 kernel: AMD Features=0x28100800<SYSCALL,NX,RDTSCP,LM>
Mar 8 02:01:32 kg-v7 kernel: AMD Features2=0x1<LAHF>

As you can see, Intel cpu runs perfectly on the amd64 branch.
Unless your Xeons are old 32-bit only, they will also rund amd64.
--
Torfinn Ingolfsen,
Norway
 | 
Pages: 1
Prev: fatal trap 12 page fault while in kernel mode
Next: 8.0 sysinstall crashes