I have an MSI K8D-Master-F motherboard with two Opteron 244's.
It recently got retasked as a desktop machine due to another
system's failure...I thought the little ATI Rage built into
the motherboard would make more than a sufficient 2d desktop
type framebuffer, and as far as other stuff inside this box...
this should be a pretty kick ass, high performance workstation.
But I can literally watch the screen redraw if I switch virtual
desktops. Painfully slow. Thinking this was just some
crippled built-in video bug, I threw in an old nVidia card
we had lying around in one of the regular PCI slots. Slightly
slower, if you can believe it.
I think this:
mtrr: type mismatch for e5000000,1000000 old: uncachable new: write-combining
And of course this:
$ cat /proc/mtrr
reg00: base=0xc0000000 (3072MB), size=1024MB: uncachable, count=1
reg01: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
Has something to do with it (and may also explain why I'm getting
less than stellar performance out of other PCI devices, such as
the RAID array that has a 128MB memory region the driver directly
reads and writes on). The above is with the nVidia PCI card
inserted, the built-in ATI card would also emit a similar mtrr
rejection message, just the address is different, and the system
would only list 512MB uncachable without the nVidia.
Also notice that there's 4GB installed in this machine, in 2 2GB
DIMMs installed in one bank together. And yet the BIOS map:
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 00000000bfff0000 (usable)
BIOS-e820: 00000000bfff0000 - 00000000bffff000 (ACPI data)
BIOS-e820: 00000000bffff000 - 00000000c0000000 (ACPI NVS)
BIOS-e820: 00000000ff7c0000 - 0000000100000000 (reserved)
Lists nothing beyond the 4GB mark where that memory might
reasonably have been moved to. Assuming this is just a
neglectful BIOS, I tried setting 'mem=4G' and 'memmap=1G@4G'
kernel options, to define the 1GB 'hole' I presume is being
placed here, and which got me this added line to dmesg:
user: 0000000100000000 - 0000000140000000 (usable)
But still only 3GB available.
To add even more insult to rising injury, I noticed while
looking through dmesg that the IOMMU is being disabled. It
appears the value being advertised by the CPU's:
CPU 0: aperture @ 1b80000000 size 128 MB
Is being rejected because it lies (WELL) above the 4GB mark.
I've been going through archives and I'm not finding anything
that looks like actionable advice.
The latest BIOS MSI's webpage lists for this motherboard is 1.1:
already installed. It has no option for 'pci hole: software'.
The only options I'm finding are memory bank interleaving (which
I tried disabling), and 'Disabled/Best-Fit/Absolute' settings
for the IOMMU, as well as its aperture size (128M, which I
might increase if it would be used rather than ignored...).
One post in the archive that matched on motherboard make and
model looks confused to me, because he refers to his BIOS as
AWARD 2.0, but the only BIOS MSI puts out for this board is
AMI (versioned 1.1).
In summary what I'd like to know is:
1) Where is the 'go fast' button? re: pci video that's slower than
even isa video should be (and presumably similar performance
problems on other PCI devices).
2) How do I get the last 25% of my memory back?
3) How do I get the system to put the IOMMU somewhere in the range
it stole under 4G (so Linux can use it)? Can Linux move this on
its own accord?
Thanks in advance for any and all help.
--
David W. Hankins "If you don't do it right the first time,
Software Engineer you'll just have to do it again."
Internet Systems Consortium, Inc. -- Jack T. Hankins
Hello,
after long frustrating days attempting to install Debian on
my new system, I gave up and returned back to SuSE 10.2.
Installation from DVD worked like a charm. (Except a small
problem with an existing RAID, which could be solved by
cleaning the discs with sand-paper :-)
(seriously: dd if=/dev/null of=/dev/sdxxx)
Now my current problem:
I would like to use a Matrox G550 in dual head mode.
At the moment, I am able to configure a single monitor
with yast2/sax2 - no problem.
According to an older posting, I would like to add an
additional Display section in XF86Config.
Now, the problem is: where is XF86Config?
(I had a look at /etc/X11 and also tried a 'find')
At the moment, the behaviour is quite strange:
A Samsung Syncmaster 204B is attached to Port A
A CRT ADI monitor is attached to Port B
The 'console' (F1) is shown on both screens.
After startup of runlevel 5, the X11 display appears
on the ADI monitor (flickering), whereas the Samsung
remains dark.
If I swap the connectors, the situation remains the same.
Any ideas/suggestions?
Thanks in advance.
Greetings Harald
--
Dr.-Ing. Harald Finster / Aachen Germany
http://www.finster-stahlart.de industrial history and architecture
http://www.astrid-aix.de gallery: watercolours and oil paintings
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
Hi ,
Can anyone help me to find the following ?
Happened to see /proc/<pid>/mapped_base entry in SLES9/10 AMD64 kernels. And it has the default value 1431654400 ( 0x55555000 ).
$ cat /proc/29270/mapped_base
1431654400
This entry was introduced in SLES and RH in the past to alter the TASK_UNMAPPED_BASE entry dynamically ( to shift the mmap starting address ). And was effective in 32 bit kernels a lot to get more range for mmaps ( or to create large shared memory area).
Not sure, why this entry is still present in SLES kernels and altering this entry doesn't help also ( ie mmap always starts at 0x2a95556000 = TASK_SIZE/3 . Can be verified with a small program and maps entry for the program).
Is it a bug ? 1) Either it needs to be removed 2) or needs to be implemented so that users can alter this value ?
Thanks in advance ,
Nagraj.
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
Hi!
We would like to build a cluster of multiple Opteron based computers.
The usage is for scientific computing (multiple scientists running
large processes usually at the same time... :-) Mainly computing but
also some large flat file access, no DB stuff.
Can this be done?
(I think it can: http://www.theinquirer.net/default.aspx?article=36371
BMW seems to have done it except with woodcrests and not Opterons).
How? Can't find anything really about getting any SUSE version or 2.6
kernel for that matter to work in cluster. Anybody have any
experience? Any pointers to resources? Opensourse? Commercial?
Whatever?
--
HG.
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
Afternoon all,
Open Suse 10.2 is now installed. I have a question about VNC sessions onto
the Suse Box using RealVNC viewer. I cannot connect to the same session
after disconnecting it like using Windows RealVNC server.
This question has popped up a lot in my google searches and there are no
answers for these guys! Is it because it is not possible? I use VNC for all
my other boxes on my network and this is my first linux machine so it would
be good if I could use the same stuff for RAS.
How do you set the VNC server in linux to accept reconnection to old
sessions? I have tried IPAddress:SessionNo ie: 192.168.0.x:1 and that always
starts a new session where you have to log in. Same with
ComputerName:SessionNo or just the computername or IPAddress.
I'm guessing there are settings to change - but alas, my skills are poo. :)
Thanks for any help.
Ben
_________________________________________________________________
Windows Live Messenger has arrived. Click here to download it for free!
http://imagine-msn.com/messenger/launch80/?locale=en-gb
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
Hello Everyone. Have SuSE 10.2 installed on my Laptop (Acer Ferrari 5000) using KDE.
The 'only' problem problem found so far (apart from waiting for the ATI-driver to be released) is that when connecting headphones the internal speakers does not mute. This is a "combined" headphone/SPDIF socket where I believe the muting is SW-controlled.
I found some information saying that Kmix should have an option called.
<Quote> Switches tab: mute 'Headphone Jack Sense' and 'Line Jack Sense' </Quote>
However I do not have any 'Switches tab' at all and wonder if this a side effect of ALSA not being able to fully handle my Sound Card.
Each time i boot the machine I see several errors (perhap 50-100 rows) like this :
ALSA sound/pci/hda/hda_codec.c:212: hda_codec: invalid dep_range_val 0:7fff
ALSA sound/pci/hda/hda_codec.c:212: hda_codec: invalid dep_range_val 0:7fff
ALSA sound/pci/hda/hda_intel.c:590: hda-intel: get_response timeout: IRS=0x10
ALSA sound/pci/hda/hda_codec.c:212: hda_codec: invalid dep_range_val 0:7fff
ALSA sound/pci/hda/hda_codec.c:212: hda_codec: invalid dep_range_val 0:7fff
ALSA sound/pci/hda/hda_intel.c:590: hda-intel: get_response timeout: IRS=0x10
ALSA sound/pci/hda/hda_codec.c:212: hda_codec: invalid dep_range_val 0:7fff
ALSA sound/pci/hda/hda_codec.c:212: hda_codec: invalid dep_range_val 0:7fff
ALSA sound/pci/hda/hda_intel.c:590: hda-intel: get_response timeout: IRS=0x10
Sound Card: 00:14.2 Audio device: ATI Technologies Inc SB450 HDA Audio (rev 01)
Just want a second opinion before filing a bug report.
TIA.
Brgds Jonas Israelsson
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
Hi!
On 12/18/06, Jan Engelhardt <jengelh(a)linux01.gwdg.de> wrote:
>
> On Dec 18 2006 11:24, HG wrote:
> >Subject: [opensuse] How to build a cluster with 64 bit Opterons and SUSE?
> >
> > Hi!
> >
> > We would like to build a cluster of multiple Opteron based computers.
> > The usage is for scientific computing (multiple scientists running
> > large processes usually at the same time... :-) Mainly computing but
> > also some large flat file access, no DB stuff.
> >
> > Can this be done?
> > (I think it can: http://www.theinquirer.net/default.aspx?article=36371
> > BMW seems to have done it except with woodcrests and not Opterons).
> >
> > How? Can't find anything really about getting any SUSE version or 2.6
> > kernel for that matter to work in cluster. Anybody have any
> > experience? Any pointers to resources? Opensourse? Commercial?
> > Whatever?
>
> It does not take much. Clusters at GWD run some SLES8 with LSF, throw
> in some MPICH programs where you need it, and you are all set.
> Or were you looking for an OpenMOSIX-style cluster? That would be
> a different matter harder to solve.
Thanks... I think :-)
Sorry, I do have to go and look for MPICH and LSF... as I do not know
what those are. OpenMosix... I think I know.
Little bit more about the needs:
- multiple users (around 12+users)
- all running scientific mathematical calculations on software ranging
from matlab to self programmed software running with java, python, R,
you name it
- programs take from few minutes to days to run, usually at least
couple of hours
- users have hard time to find a "free" CPU/memory
So, if we had the calculation servers set up as a cluster, users could
find computing resources easily. Or more to the point, they would not
have to do that, they just login to the (cluster) server and let the
server worry about where it's running their processes.
So, I do not know if your terms apply here or not. Altough, I think
MPICH sounds like specialized software which we do not have (in that
sense at least).
--
HG.
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
I just installed opensuse 10.2 on my Asus K8V-X SE with AMD Athlon 1800
running at 1533 MHz, 768MB, 120G hard drive. I have my drive
partitioned with 2G swap, 2G /var, 20G /, and 87G /home. Usually I only
burn CD1, abort the installation, and finish installation with the DVD
iso in my home directory. This time, CD1 corrupted my grub, so I had to
use another computer to burn the other two CD's and reinstall.
Things seemed to go well until I logged onto the system after
installation, whereupon the crash dialog box came up saying the front
panel (kicker) had crashed with a signal 8, and I should notify the
appropriate people of a possible bug in the application. All the KDE
window manager stuff seems to have disappearted, but X runs fine, and I
had a kterm running already, from which I can start, for instance,
Konqueror, Firebird, and Thunderbird (which I'm using for this email).
I have no application menu, and only the initial basic icons on the
single desktop that seems to have survived the crash.
I saved the detailed message tab text, and it's attached, in case it can
help. How can I get KDE back up again?
--
John Perry
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
Today I received my 10.2 x64 DVD from Cheapbytes. When I tried to
upgrade from my 10.0 to 10.2, the dependency checker seemed to hang.
Namely, after the dependency window popped, I tried to resolve all
dependencies manually. Eventually, I believe I did. Then, after
hitting "Accept" to proceed with the upgrade, an ***EMPTY***
dependencies window popped up. Since it showed no dependencies or
conflicts, I hit "Cancel" and then "Accept" in the YAST window to
proceed. The dependencies window popped up again, empty. From there
on, I got stuck in this loop.
My original 10.0 x64 was the retail version. Since installing it, I
updated several packages via apt-get and smart. (Some of them appeared
locked in the YAST window; in one of my iterations, I manually unlocked
them, but that did not help as I was caught in the same loop as
described above.) The DVD I just got is Opensuse. Could that create a
problem? Will I have to install 10.0 from scratch? Should I upgrade to
10.1 first and then 10.2?
TIA for your time.
--
Running 64-bit Linux on AMD64
---------------------------------------------------------------------
To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org
New to Suse Linux,Should I install 32 bit or 64 bit version.
PLEASE ADVICE
I have AMD Athlon(tm) 64 Processor 3200+ Socket 754
Asrock K8NF4G-SATA2
Nvidia 6600LE 256MB PCI-E
512X2(1GB) Ram
80GB sata hdd
are all apps supported under 64 bit like beryl project etc.Also what is the maximum space required to install linux
Is 20 GB enough for full installation(programming ,server) and add on apps. -oswin
_________________________________________________________________
Windows Live Spaces is here! It’s easy to create your own personal Web site.
http://spaces.live.com/?mkt=en-in