Erez Schatz | 1 Jul 01:40 2006
Picon

Re: Support for ThinkPad 390X?

Sorry to bump this, but, anyone has any ideas?

On 6/28/06, Erez Schatz <moonbuzz <at> gmail.com> wrote:
Hello,
I've been using an old ThinkPad 390X as a study machine, installing different OS's and testing, and I've came across Plan9.
I've tried the LiveCD and while installation seems fine, I can't seem to be able to run it, as it gets stuck in mid-boot.
I didn't find it in the supported hardware (or maybe didn't look hard enough...)
Question is, is the 390X supported by Plan9, and if so, any ideas on what isn't going right or how to run it?

Thanks in advance

--
Erez

I debug, therefore I code.
A Blog of Very Little Brain: http://moonbuzz.blogspot.com



--
Erez

I debug, therefore I code.
A Blog of Very Little Brain: http://moonbuzz.blogspot.com
Rodolfo (kix | 1 Jul 02:45 2006
Picon

Re: Re: Support for ThinkPad 390X?

Hi,

I dont have any ideas about why your X390 have a problem. I try with one and all is ok.

2006/7/1, Erez Schatz <moonbuzz <at> gmail.com>:
Sorry to bump this, but, anyone has any ideas?


On 6/28/06, Erez Schatz <moonbuzz <at> gmail.com > wrote:
Hello,
I've been using an old ThinkPad 390X as a study machine, installing different OS's and testing, and I've came across Plan9.
I've tried the LiveCD and while installation seems fine, I can't seem to be able to run it, as it gets stuck in mid-boot.
I didn't find it in the supported hardware (or maybe didn't look hard enough...)
Question is, is the 390X supported by Plan9, and if so, any ideas on what isn't going right or how to run it?

Thanks in advance

--
Erez

I debug, therefore I code.
A Blog of Very Little Brain: http://moonbuzz.blogspot.com



--
Erez

I debug, therefore I code.
A Blog of Very Little Brain: http://moonbuzz.blogspot.com



--
Rodolfo García "kix"
geoff | 1 Jul 06:17 2006
Picon

Re: Re: Any one going to change to sata disks?

The fs kernel's sdata.c hadn't been recognising Silicon Image's and
other VIDs and DIDs recognised by the current cpu kernel's sdata.c
I've just updated source's /sys/src/fs/pc/sdata.c to incorporate the
atapnp() code from the cpu kernel to recognise Silicon Image SATA,
Nvidia, ATI, Serverworks and other controllers.

quanstro | 2 Jul 01:58 2006
Picon

Re: AMD Sempron

only socket A semprons are k7s.
from AMD's site the 3300+ socket 754 is stepping D0, E3 or E6.

the E3/E6 is a 64-bit Palermo core with mmx/sse2/3dnow! with a 16bit 800Mhz HT link.
the D0 is the same but might (!) not support 64bit and the cache sizes are half as bug.
(64/128 vs. 128/256).

- erik

http://www.amdcompare.com/us-en/desktop/Default.aspx
http://www.cpu-world.com/CPUs/K7/index.html
http://www.cpu-world.com/CPUs/K8/TYPE-Sempron%2064.html

Lyndon Nerenberg | 2 Jul 02:34 2006
Picon

Re: AMD Sempron

> the E3/E6 is a 64-bit Palermo core with mmx/sse2/3dnow! with a  
> 16bit 800Mhz HT link.
> the D0 is the same but might (!) not support 64bit and the cache  
> sizes are half as bug.
> (64/128 vs. 128/256).

Bloody hell.  Let's just simplify the whole mess and give them all a  
brand name of CPU™.

I've have a new plan:  I'll buy the Sempr0n box to use as a  
replacement for my BSD server, and run Plan 9 on the old box (a Via  
C3 Nehemiah).  This works better anyway, as the Via "server" will  
actually fit on the boat ;-)

--lyndon
quanstro | 2 Jul 02:48 2006
Picon

Re: AMD Sempron

it's not that bad.  they're all K8s with different steppings.  what you don't 
know is the cache size or if it's got 64-bit capabilities, which kenc does not
yet support.  i don't think what you don't know will be a problem in this
case.

it's interesting with all the marketing going on right now you need to be
quite savvy to figure out what you're really getting.  a bit like the days of
hand-built computers.

- erik

On Sat Jul  1 19:35:43 CDT 2006, lyndon <at> orthanc.ca wrote:
> > the E3/E6 is a 64-bit Palermo core with mmx/sse2/3dnow! with a  
> > 16bit 800Mhz HT link.
> > the D0 is the same but might (!) not support 64bit and the cache  
> > sizes are half as bug.
> > (64/128 vs. 128/256).
> 
> Bloody hell.  Let's just simplify the whole mess and give them all a  
> brand name of CPU™.
> 
> I've have a new plan:  I'll buy the Sempr0n box to use as a  
> replacement for my BSD server, and run Plan 9 on the old box (a Via  
> C3 Nehemiah).  This works better anyway, as the Via "server" will  
> actually fit on the boat ;-)
> 
> --lyndon

jmk | 2 Jul 04:19 2006

Re: AMD Sempron

yes. you can fuss over the micromanagement that
ekes out every fractional % of performance but unless
you are in a very specialised environment that's
just obsession. the chips either implement the
instruction set and architectural features you expect
or they don't. plan9 expects little from the cpu.

as usual, what's round about the cpu is the problem,
graphics, network, and now to some extent usb (and
sata?) being the primary concerns for support. but,
of course, device drivers aren't hard (are they? if
they were why do other operating systems have so many,
coded by such obviously mediocre programmers?), so a
little elbow grease should fix that.

On Sat Jul  1 20:59:23 EDT 2006, quanstro <at> quanstro.net wrote:
> it's not that bad.  they're all K8s with different steppings.  what you don't 
> know is the cache size or if it's got 64-bit capabilities, which kenc does not
> yet support.  i don't think what you don't know will be a problem in this
> case.
> 
> it's interesting with all the marketing going on right now you need to be
> quite savvy to figure out what you're really getting.  a bit like the days of
> hand-built computers.
> 
> - erik
> 
> On Sat Jul  1 19:35:43 CDT 2006, lyndon <at> orthanc.ca wrote:
> > > the E3/E6 is a 64-bit Palermo core with mmx/sse2/3dnow! with a  
> > > 16bit 800Mhz HT link.
> > > the D0 is the same but might (!) not support 64bit and the cache  
> > > sizes are half as bug.
> > > (64/128 vs. 128/256).
> > 
> > Bloody hell.  Let's just simplify the whole mess and give them all a  
> > brand name of CPU™.
> > 
> > I've have a new plan:  I'll buy the Sempr0n box to use as a  
> > replacement for my BSD server, and run Plan 9 on the old box (a Via  
> > C3 Nehemiah).  This works better anyway, as the Via "server" will  
> > actually fit on the boat ;-)
> > 
> > --lyndon

Lyndon Nerenberg | 2 Jul 07:18 2006
Picon

Re: AMD Sempron

> you need to be
> quite savvy to figure out what you're really getting.  a bit like  
> the days of
> hand-built computers.

When I build them by hand I *know* what I'm getting.  That's the  
problem here :-(

Anyway, if the little bugger will play nice as a 32-bit CPU I suppose  
I'm okay.  But that BSD machine could use a kick in the pants, so I  
think I'll stick to the plan, anyway.

And I don't even want to ask about the G5 box I'm working on getting  
my fingers on.

--lyndon

Lyndon Nerenberg | 2 Jul 07:22 2006
Picon

Re: AMD Sempron

> the chips either implement the
> instruction set and architectural features you expect
> or they don't.

Not having played with any of the 64 bit intel/amd gear I'm not  
really comfortable with having to figure out the hard way just how  
"compatible" they are with 32 bit mode, despite (or in spite of)  
claims made by the manufacturers.  Hardware is just too fscking  
weirde these days.

--lyndon

quanstro | 2 Jul 09:43 2006
Picon

Re: AMD Sempron

they're 100% compatable in 32bit mode.

On Sun Jul  2 00:23:10 CDT 2006, lyndon <at> orthanc.ca wrote:
> > the chips either implement the
> > instruction set and architectural features you expect
> > or they don't.
> 
> Not having played with any of the 64 bit intel/amd gear I'm not  
> really comfortable with having to figure out the hard way just how  
> "compatible" they are with 32 bit mode, despite (or in spite of)  
> claims made by the manufacturers.  Hardware is just too fscking  
> weirde these days.
> 
> --lyndon


Gmane