August 2003 - openSUSE AMD64 - openSUSE Mailing Lists

Tyan S2880 244 6GB
by Santiago Flores 30 Sep '03

30 Sep '03

Hello, I downloaded and installed Suse 8.2Beta. Straightforward install. No issues (other than having to turn off APIC - same for all other distros). The machine was up for about a week and a power problem required it to be shutdown. Now when I reboot the whole process hangs at the intialization of cpu #1. Any ideas? I booted with a rescue disk to checkout the disks and they were fine. I booted with a ClusterKnoppix CD that I had laying around and it worked fine, with both CPUs. Attached is the serial output of a boot with no APIC. Help please

5 7

Re: [suse-amd64] Installing directly to SATA drive on Asus SK8N
by Martin Jungowski 31 Aug '03

31 Aug '03

No, you're trying the right thing. Have you tried not compiling the nvaudio driver at all and just make make install in the nvnet directory? That way you can avoid nvaudio and nvgart which ain't working anyway Martin On Sun, 2003-08-31 at 17:03, Phil Chapman wrote: > ok, > > I'd already tried -mathlon -m64 and the -march still causes a problem. > Taking out -mathlon gets past there but there are many compile errors. > > Looking in the mobo manual though, the LAN is described as a > > CK8 MAC+ Realtek RTL8201BL PHY > > Is this nvidia?? I'm not so sure, so we might be trying the wrong thing > anyway. > > Phil > > > On Sun, 31 Aug 2003, Martin Jungowski wrote: > > > Hi, > > > > Alright, I found the following in the GCC manual: > > > > -m64 > > Generate code for a 32-bit or 64-bit environment. The 32-bit > > environment sets int, long and pointer to 32 bits and generates > > code that runs on any i386 system. The 64-bit environment sets > > int to 32 bits and long and pointer to 64 bits and generates > > code for AMD's x86-64 architecture. > > > > First, simply try to add the -m64 parameter to the CFLAGS options and > > use "-march=athlon". If that doesn't work either, delete the -march > > parameter and use -m64 only. > > > > Martin > > > > > > > > On Sun, 2003-08-31 at 16:25, Phil Chapman wrote: > > > hmmm, > > > > > > uname -m returns x86_64 > > > > > > uname -p returns unknown > > > > > > I changed the CFLAGS to be -march=x86_64 > > > > > > but it gets an unknown value for march when doing the make. > > > > > > tried -march=i686 and again it gets CPU does not cuppot x86_64 instruction > > > set?? > > > > > > Any more ideas?? > > > > > > How can I tell if I'm in 64-bit mode? > > > > > > Regards > > > > > > Phil > > > > > > > > > >

2 2

Re: [suse-amd64] Installing directly to SATA drive on Asus SK8N
by Martin Jungowski 31 Aug '03

31 Aug '03

Hey Phil, Whether the onBoard NIC is going to work or not depends on the revision of the mainboard. We got two of'em for testing and debugging, Rev. 1.02 is what I'm using right now and Rev 1.03 which another of our hardware testers is going to use starting next week (also getting a new AMD CPU :)). I've had no trouble activating it on 1.03, what drivers are you using? We got Beta drivers from nVidia which seem to work just fine, however I am not allowed to give that driver away and it's not yet available to public :( However I've had a very similar problem with 1.02: I's able to load the module and I was also able to ping my server and several websites. I was not able to ssh to my server or to load any of the websites in any webbrowser. No trouble whatsoever with 1.03, what Revision do you have? You should have 1.03 since 1.02 wasn't available to public either and is still a prototype. SuSE 8.2/AMD64 does not have the proper driver for the onBoard NIC yet, you'd have to use the nVidia driver. But first, could you be so kind and tell me what revision of the SK8N you have, 1.02 or 1.03? What version is your BIOS? Maybe there's something we can do about that onBoard NIC ;) Martin On Sun, 2003-08-31 at 00:47, Phil Chapman wrote: > Hi Martin, > > thanks for the tips. I've just installed it on an 80Gb WD. No problems > except for the network. Have you managed to get the on-board LAN working? > > It seemed to detect it during install, but then failed when doing it's > test connect :( > > Any ideas? At this stage I was just about to put in a different nic. > > It seems like a great board though!!! > > Thanks > > Phil > > > > On Sun, 31 Aug 2003, Martin Jungowski wrote: > > > Hi, > > > > You're lucky cause I got that exact same mainboard & CPU and I've been > > running SuSE x86-64 for a couple of weeks now ;) > > There are Drivers available for this S-ATA controller but I haven't > > tested them yet: > > > > http://www.busybox.net/pdc-ultra-1.00.0.10.tgz > > > > It should be working though, give it a shot. However I'd suggest you > > install SuSE x86-64 on a P-ATA drive connected to the onBoard onChipset > > IDE-Controller. Don't forget to enable UDMA and 32-bit mode for the > > harddrive afterwards: > > > > hdparm -d1 -c1 /dev/hda > > > > Note that this will only work with the SuSE default kernel 2.4.19 but > > not with any other self compiled kernel yet. Apparantly, SuSE updated > > the IDE-driver in that specific kernel and those patches are not part of > > 2.6 yet. > > > > Martin > > > > > > > > On Sat, 2003-08-30 at 20:20, Phil Chapman wrote: > > > Hi, > > > > > > just subscribed to the list. I got an opteron 144 on an Asus SK8N mobo > > > this week with a 120Gb seagate SATA drive. > > > > > > Can SuSE 8.2 be installed directly to the drive ? I can't get it to > > > recognise it at all, but am not sure which drivers to use? Or should I > > > just install to a default IDE drive and use the SATA drive seperately? > > > > > > Thanks for any help > > > > > > Phil Chapman > > > > > >

2 9

Installing directly to SATA drive on Asus SK8N
by Phil Chapman 30 Aug '03

30 Aug '03

Hi, just subscribed to the list. I got an opteron 144 on an Asus SK8N mobo this week with a 120Gb seagate SATA drive. Can SuSE 8.2 be installed directly to the drive ? I can't get it to recognise it at all, but am not sure which drivers to use? Or should I just install to a default IDE drive and use the SATA drive seperately? Thanks for any help Phil Chapman

3 2

SK8N / nForce3 trouble
by Martin Jungowski 30 Aug '03

30 Aug '03

Hi, Marc Miller with AMD sent me here 'cause I'm one of the lucky owners of an AMD Opteron computer. I have an SK8N with nForce3 chipset and AMD Opteron 144 processor and I'm willing to do everything I can to help developing drivers and fixing problems with that platform ASAP, don't care whether it's SuSE or RedHat or Debian or whatever. Right now I'm having some UDMA trouble, I cannot set the DMA flag via hdparm in 32-bit legacy mode, running SuSE 8.2. I have no trouble at all activating it in SuSE 8.2 x86-64 for some reason. FYI, same issues with Debian and Kernel 2.4.21, 2.4.22, 2.6.0-test3 and 2.6.0-test4 but I can set the UDMA-flag using the Debian standard kernel-image 2.4.18-1-k7. I don't really understand what the problem is, I'd guess the chipset is not fully supported yet but that ain't no explanation why - it's running just fine in 64-bit mode with SuSE 8.2 x86-64 Beta - it's running just fine with Debian's kernel-image-2.4.18-1-k7 I have not found a way to enable DMA in 32-bit mode in SuSE yet. I'm having some more trouble too but that's not SuSE related, so I'm not gonna bother you with that now, like the ECC trouble. Thanks for your help and please tell me if there's anything I can do to help developing and improving the 64-bit port. Martin

2 5

Re: [suse-amd64] SK8N / nForce3 trouble
by Andi Kleen 30 Aug '03

30 Aug '03

On Sat, 30 Aug 2003 01:44:57 +0200 Martin Jungowski <jungowsk(a)in.tum.de> wrote: > > And ACPI works for you? > > > It does not work with the standard kernel and in order to install SuSE > 8.2 (either i386 or x86-64), I have to disable ACPI, APIC and USB in > BIOS Setup and also I have to use "pci=noacpi" and "noapic". Same with Just acpi=off doesn't work? -Andi

2 1

RE: [suse-amd64] SK8N / nForce3 trouble
by david.obrien＠amd.com 30 Aug '03

30 Aug '03

> > It does not work with the standard kernel and in order to install SuSE > > 8.2 (either i386 or x86-64), I have to disable ACPI, APIC and USB in > > BIOS Setup and also I have to use "pci=noacpi" and "noapic". Same with > > Just acpi=off doesn't work? Is there hope for a SLES8 kernel that does not require this?

1 0

Failed to compile 10Gb module
by Dan 28 Aug '03

28 Aug '03

Hi List, SuSE 8.2 beta runs fine until I tried to compile Intel 10Gb driver source on a dual Opteron server ;-(. Following errors occurred from "make install". Any suggestions? depmod: *** Unresolved symbols in /lib/modules/2.4.19-SMP/pcmcia-external/avm_cs.o depmod: b1pcmcia_addcard_b1 depmod: b1pcmcia_delcard depmod: b1pcmcia_addcard_m1 depmod: b1pcmcia_addcard_m2 depmod: *** Unresolved symbols in /lib/modules/2.4.19-SMP/pcmcia-external/avma1_cs.o I saw 2.5/2.6 kernel now has a newer Intel 10Gb driver support. When I tried to compile 2.5 kernel gcc gave me a bunch of errors in "make modules". Has anyone tried to run 2.5/2.6 with this beta SuSE? Thanks very much for your input. -Dan __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

2 1

Tyan S2880 244 6GB
by Santiago Flores 27 Aug '03

27 Aug '03

Ok. It is booting now with IOMMU aperture at 2GB, APIC off (in bios) and mem=4G. The system now reports an incorrect amount of ram in /proc/meminfo: mojon:~ # cat /proc/meminfo total: used: free: shared: buffers: cached: Mem: 1899364352 238714880 1660649472 0 22364160 79704064 Swap: 4573167616 0 4573167616 MemTotal: 1854848 kB MemFree: 1621728 kB MemShared: 0 kB Buffers: 21840 kB Cached: 77836 kB SwapCached: 0 kB Active: 53296 kB Inactive: 46472 kB HighTotal: 0 kB HighFree: 0 kB LowTotal: 1854848 kB LowFree: 1621728 kB SwapTotal: 4465984 kB SwapFree: 4465984 kB BigFree: 0 kB Ideas? Also, it was booting before with 6GB without passing anything to the command line. What could have changed/because of the power failure? Nothing comes to mind right away. Thanks for all of the help getting the system booting. Santiago

2 1

Re: [suse-amd64] Tyan S2880 244 6GB
by Andi Kleen 27 Aug '03

27 Aug '03

On Wed, 27 Aug 2003 10:11:37 -0700 "Santiago Flores" <santi(a)mleads.com> wrote: > > I did attempt to boot with mem=4G and got the attached output. Very > interesting. The MCE error points to some problem in the memory. try switching DIMMs around to other slots or replace them. You can turn it off with mce=off (that is what 32bit kernels do by default), but that would only hide the problem. > > I have upped my IOMMU to 2GB and am playing with APIC settings to see if I > can get it to boot. That's not a good idea. You will lose memory this way with 6GB. Keep it at 128-256MB. -Andi

2 1

RE: [suse-amd64] Tyan S2880 244 6GB
by Miller, Daniel J. 27 Aug '03

27 Aug '03

How severe was the "power problem?" I've had to re-install earlier Linux's after an accidental shutdown on a running system (wasn't on an Opteron system, but this sounds like a Linux filesystem issue rather than an AMD64 issue). I haven't seen it happen yet on SuSE 8.2, but I haven't seen many accidental power-offs on 8.2, either. It's one area in which NTFS seems to be a ahead of ext3 and reiserfs. Is the system on a UPS? Dan -----Original Message----- From: Santiago Flores [mailto:santi@mleads.com] Sent: Wednesday, August 27, 2003 11:05 AM To: suse-amd64(a)suse.com Subject: [suse-amd64] Tyan S2880 244 6GB Hello, I downloaded and installed Suse 8.2Beta. Straightforward install. No issues (other than having to turn off APIC - same for all other distros). The machine was up for about a week and a power problem required it to be shutdown. Now when I reboot the whole process hangs at the intialization of cpu #1. Any ideas? I booted with a rescue disk to checkout the disks and they were fine. I booted with a ClusterKnoppix CD that I had laying around and it worked fine, with both CPUs. Attached is the serial output of a boot with no APIC. Help please

1 0

RE: [suse-amd64] System freeze with NVIDIA 1.0-4180 on AMD64
by david.obrien＠amd.com 26 Aug '03

26 Aug '03

> Has anyone had any luck with the NVIDIA 1.0-4180 drivers on an AMD64 > system? Yes I have successfully used the driver with SLES 8. -- David

2 1

Re: [suse-amd64] Problem with strange CPU priority.
by Andreas Jaeger 23 Aug '03

23 Aug '03

On Sat, Aug 23, 2003 at 06:42:28PM +0200, Daniel Persson wrote: > Hi Andreas, > [...] > > You can bind processes to specific cpus via the /proc/interface or the > > numactl program. > > Ok, ill try that, but first, i downloaded the irqbalance program and have > it running on all machines. Before irqbalance /proc/irq/xx/smp_affinty > looked like this: > > > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > > And after it looked like this. That's fine. > 00000002 > 00000001 > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > 00000001 > 00000002 > 00000002 > ffffffffffffffff > 00000001 > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > ffffffffffffffff > 00000001 > (after running cat /proc/irq/*/smp_affinity > > Also a quick check with cat /proc/interrupts > CPU0 CPU1 > 0: 8578447 249920 IO-APIC-edge timer > 1: 15 0 IO-APIC-edge keyboard > 2: 0 0 XT-PIC cascade > 9: 794975248 0 IO-APIC-level acpi > 14: 35830 1 IO-APIC-edge ide0 > 15: 28 1 IO-APIC-edge ide1 > 19: 0 0 IO-APIC-level usb-ohci, usb-ohci > 24: 2288256 0 IO-APIC-level eth0 > NMI: 8575636 5437943 > LOC: 8829221 8829220 > ERR: 0 > MIS: 0 > > Show that CPU1 (this is right after running iqrbalance - so i guess that > it should change over time) have not handled near as much interruppts as > CPU0. Is my reading of this correct ? Is this normal ? The numbers are calculated since system start. The load is evenly distributed soon but not in the way that the same interrupte, e.g. for eth0 is handled by both cpu0 and cpu1 but in the way that CPU0 and CPU1 get in sum the same number of interrupts. So, cpu0 will now handle e.g. timer, acpi, eth0 and cpu1 will handle ide0, ide1 and timer. This also might give besser locality. > > > > > Are you using 2 2p boards in each case? Or a 4p one? > Both setups are with 2p boards. Ok. > > One question, im not near the machines right now, but, i they where > deliverd with all memory installed on CPU0, and no modules at CPU1, could > this generate this kind of problems ? It shouldn't ;-). For performance reasons I would change it. > (i could imagnie that it could) Andreas -- Andreas Jaeger, aj(a)suse.de, http://www.suse.de/~aj SuSE Linux AG, Deutschherrnstr. 15-19, 90429 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126

2 1

[Fwd: Re: [suse-amd64] Problem with strange CPU priority.]
by Daniel Persson 23 Aug '03

23 Aug '03

Hi Andreas, > Replace it with a machine running SuSE ;-). Seriously, this shouldn't > be a problem. Nope, it should not, but this is the most obvius difference in the two setups, so i tought i mention it ;) > You can bind processes to specific cpus via the /proc/interface or the > numactl program. Ok, ill try that, but first, i downloaded the irqbalance program and have it running on all machines. Before irqbalance /proc/irq/xx/smp_affinty looked like this: ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff And after it looked like this. 00000002 00000001 ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff 00000001 00000002 00000002 ffffffffffffffff 00000001 ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff 00000001 (after running cat /proc/irq/*/smp_affinity Also a quick check with cat /proc/interrupts CPU0 CPU1 0: 8578447 249920 IO-APIC-edge timer 1: 15 0 IO-APIC-edge keyboard 2: 0 0 XT-PIC cascade 9: 794975248 0 IO-APIC-level acpi 14: 35830 1 IO-APIC-edge ide0 15: 28 1 IO-APIC-edge ide1 19: 0 0 IO-APIC-level usb-ohci, usb-ohci 24: 2288256 0 IO-APIC-level eth0 NMI: 8575636 5437943 LOC: 8829221 8829220 ERR: 0 MIS: 0 Show that CPU1 (this is right after running iqrbalance - so i guess that it should change over time) have not handled near as much interruppts as CPU0. Is my reading of this correct ? Is this normal ? > > Are you using 2 2p boards in each case? Or a 4p one? Both setups are with 2p boards. One question, im not near the machines right now, but, i they where deliverd with all memory installed on CPU0, and no modules at CPU1, could this generate this kind of problems ? (i could imagnie that it could) /Daniel

1 0

Problem with strange CPU priority.
by Daniel Persson 23 Aug '03

23 Aug '03

Hi list, I have a really strange problem, i have two different setups, one with Arima motherboards and one with Tyan motherboards. Both are running SLES8 and kernel k_smp-2.4.19-249 with all YAST2 updates. The code im running is Fluent 6.0.18 in parallell mode.(mpich) (CFD code) The problem is, that the load is very different when running in parallell, on the Arima based cluster, if running on 4CPUS (2dual machines) all cpus are occupied at 80-99% (wich should be considerd normal). However, running the exact same code and test case on the Tyan based machines(same kernel, network etc) then 2CPU is working hard and the other 2 are "watching" and this is one CPU per machine, ie one cpu in the dual machine is working, and CPU2 are idling. The result is that the same code and case are about 10times slower on the Tyan machines. (wich have the latest BIOS) The only(important?) difference in the setup is that the Tyan based test are connected to a RedHat 8.0 boxwith Dual Xeons. i.e auth and nfs are from this machine. But all computations are done on the Opteron machines. So, basically, my question are, could i have done some serious config mistake, or is there any problem with the kernel/Tyan bios wich casues one CPU to take alot more load than the other. Can i config the CPU routing somehow ? On the tyan machine, it also takes long time to just read the casefile.(ide disks, DMA and 32bitaccess enabled, same as Arima setup) Any ideas anyone ? /Daniel PS The same problems comes when using other codes aswell, such as LS-Dyna.

4 4

run-crons using an entire CPU
by Fred P. Arnold 22 Aug '03

22 Aug '03

Hello, I have a run-cron job which is executing the default scripts supplied with 8.2beta9 for AMD-64, on a Asus-based dual Opteron 240 system, which is using 100% of one of the CPUS. Top reports 6174 root 39 19 0 0 0 R 99.9 0.0 388:54.99 run-cron Any suggestion what's eating the time? -fred Frederick P. Arnold, Jr. NUIT, Northwestern U. f-arnold(a)northwestern.edu

2 1

glade-0.6.4 for SLES8-AMD64
by Stephen Williams 20 Aug '03

20 Aug '03

I had a need for this version of glade (which can co-exist with glade-2 which seems to be the new standard) so I used the src.rpm from the SuSE-8.2 Beta 9 disk. I've placed the rpms here: <ftp://ftp.icarus.com/pub/steve/x86-64/packages> There are some noarch.rpm prerequisites from the Beta9 disks that install cleanly, so get them from the disks. -- Steve Williams "The woods are lovely, dark and deep. steve at icarus.com But I have promises to keep, http://www.icarus.com and lines to code before I sleep, http://www.picturel.com And lines to code before I sleep."

1 0

System freeze with NVIDIA 1.0-4180 on AMD64
by Roderick Constance 19 Aug '03

19 Aug '03

Has anyone had any luck with the NVIDIA 1.0-4180 drivers on an AMD64 system? I have a SLES8-based sytem thas seems to have problems with an NVIDIA configuration. I can compile and load the 1.0-4180 NVIDIA_kernel and NVIDIA_GLX modules without problems. X starts up fine. But the system will lock up shortly after I start an OpenGL screensaver and move its window around. The X log and mtrrs look fine. Any suggestions? Thanks! -Roderick

1 0

AbiWord for SLES8
by Stephen Williams 19 Aug '03

19 Aug '03

Has anybody been able to build the latest AbiWord rpms for SLES8-amd64? I made a cursery attempt and got compiler errors even after I figured out the gnome-config paths. -- Steve Williams "The woods are lovely, dark and deep. steve at icarus.com But I have promises to keep, http://www.icarus.com and lines to code before I sleep, http://www.picturel.com And lines to code before I sleep."

2 6

rpm bugs on SLES8
by Stephen Williams 16 Aug '03

16 Aug '03

This is actually a distribution problem and not an AMD64 problem, but it is an AMD64 distribution, ... Anyhow, my source rpms for Icarus Verilog cause lots of warnings when I use "rpm --rebuild iverilog-0.7.20030815-0.src.rpm" to rebuild the package. Warnings like: I call /usr/sbin/Check... + /usr/sbin/Check rm: cannot remove `/usr/share/info/dir.bak.gz': Permission denied gzip: /usr/share/info/dir.bak.gz: Permission denied -rw-r--r-- 1 root root 503 2003-07-17 07:06 /usr/share/info/dir.bak.gz rm: cannot remove `/usr/share/man/man1/gimp-1.2.1.gz': Permission denied gzip: /usr/share/man/man1/gimp-1.2.1.gz: Permission denied -rw-r--r-- 1 root root 5318 2003-07-09 06:53 /usr/share/man/man1/gimp-1.2.1.gz In fact, it appears that the subcommands are not using the BUILD_ROOT that is there to allow for non-root rebuilds of packages. It may be that an old version of rpm is being used. The srpms in question is here: <ftp://ftp.icarus.com/pub/eda/verilog/snapshots/precompiled/verilog-0.7.2003…> -- Steve Williams "The woods are lovely, dark and deep. steve at icarus.com But I have promises to keep, http://www.icarus.com and lines to code before I sleep, http://www.picturel.com And lines to code before I sleep."

2 6

Re: [suse-amd64] time/date changes independently
by Ozerov Sergey 14 Aug '03

14 Aug '03

And again.... Yesterday I get nForce3 board. Install all hardware, boot to BIOS, set date, time and other options... Boot SLES and get 17 nov 2002 10:09:01 :(((( __________ www.newmail.ru -- Новая Почта: все по-новому.

1 0

bug(?) and avague initial benchmarks
by suseamd64＠rgare.com 14 Aug '03

14 Aug '03

hello list -- not sure if this is the right place to point this out, but in case it is: /etc/ssh/sshd_config Subsystem sftp /usr./lib/ssh/sftp-server s/b Subsystem stftp /usr/lib64/ssh/sftp-server also, a preliminary java single thread (thus single proc?) benchmark came back with an 18% improvment on a 244 proc oerver a xeon 2.6. no additional info atm, but it seems to bode well imo, especially considering we have more optimizing to do and also that 1.4.2 will bring about 64-bit support to the jvm...@!!! when we get more benchmark results, i'll be sure to get more detail. thanks, -- john

2 1

Amd Athlon 64 Availability
by Filipe Joel Almeida 13 Aug '03

13 Aug '03

Hi, Does anybody know when will AMD Athlon 64 be available? Also, can you tell me if all versions of Suse support 64 bits, or only the server versions? What are the best gains from using a 64 bit architecture? Thanks in advance, Filipe Joel de Almeida

3 2

Re: [suse-amd64] time/date changes independently
by Ozerov Sergey 12 Aug '03

12 Aug '03

Next try... booted sles64 - time is 08:25 open KDE "ajust date/time" and just press ok... now is 15:26 and UTC instead PDT (time zone - not selected) reboot BIOS - 15:26 boot 64 bit sles time is 08:00 PDT (KDE date applet) jusk press ok.... now is 15:01 UTC reboot BIOS - 15:02 boot 32 bit sles time is 08:03 PDT reboot boot 64 bit sles time is 06:39.... :(((( __________ www.newmail.ru -- бесплатная почта, бесплатный хостинг.

1 0

time/date changes independently
by sergozerov＠hotmail.ru 11 Aug '03

11 Aug '03

I running SLES i386 and AMD64 on Rioworks and MSI motherboards with AMD Opteron. If I set date/time on booted system and then reboot, I have date changed (for example form 9 Aug to 20 May). When running WinXP on same system(s) all works fine. It's bug or feature? :) __________ www.newmail.ru -- всегда что-то новое.

4 5

8.2 x86-64 and vmware 4.x?
by Rich McClellan 09 Aug '03

09 Aug '03

Hello und Servus! I would like to install VMWare v4.0 on a linux box (dual opteron) running SuSE v. 8.2. The VMWare installation script chokes on the differences between the running kernel and the header files. Are the header files for the SMP kernel available? thanks, Rich

4 5

Re:[suse-amd64] newbie questions
by suseamd64＠rgare.com 09 Aug '03

09 Aug '03

thanks for the help Andreas... > SLES8 ? Or that old 8.1? SuSE SLES-8 (AMD64) VERSION = 8.1 > There's no 64-bit Java out there (except GCJ which is part of GCC). Does this mean that at this moment in time there is no advantage to compiling java apps for 64-bit w/ the GCJ since the JVM that runs them is still 32 bit? > Hope you get some good numbers - feel free to share them here us down in systems are just planning on running some simple tests, but the in-house-app devs have a whole slew of tests they're gonna run, so i'll round up the results and post em here :) thanks, -- john

2 1

newbie questions
by suseamd64＠rgare.com 08 Aug '03

08 Aug '03

hello list :D we just got an opteron 244 server we're going to be using for benchmarking an in-house app against other architectures... we're running suse 8.1 i'll cut to the chase: 1) is the default compile/build behavior to take advantage of amd64 (ie: if i d/l the apache tarball and compile it, do i need to do anything special?) 2) is the default jvm modified to take advantage of amd64, or do i have to wait for the linux/jvm team to release their version of 1.4.2, or something else? (reported build is: 1.3.1, J2RE 1.3.1 IBM build cxia32131-20020622) that's it for now, but i'll prolly have more later... sorry if the questions are a little low-brow, but we all gotta start somewhere :) TIA -- john

2 1

AMD64 vs. wine
by Stephen Williams 07 Aug '03

07 Aug '03

I have the occasional Windows software that I would like to run on my Opteron. I use WINE under Linux/Intel, but it seems like I might be able to get WINE to work on Linux/x86-64. Is this a reasonable hope? -- Steve Williams "The woods are lovely, dark and deep. steve at icarus.com But I have promises to keep, http://www.icarus.com and lines to code before I sleep, http://www.picturel.com And lines to code before I sleep."

2 1

top command over 4GB.
by Yoshiki YAMAGUCHI 06 Aug '03

06 Aug '03

Hello and guten Tag, I have tested the HPL (High-Performance Linpack) benchmark on dual Opteron 240 with 8GB memory. SuSE 8.2 is very good because I could compile the benchmark and run it using memory over 3GB per process (x86's limitation). Then I noticed that the top command show false information when a process used memory over 4GB. --------------------------------------------------------------------------- top - 17:40:34 up 1 day, 54 min, 3 users, load average: 0.99, 0.97, 0.91 Tasks: 54 total, 2 running, 52 sleeping, 0 stopped, 0 zombie Cpu(s): 50.0% user, 2.1% system, 0.0% nice, 47.9% idle Mem: 8070816k total, 7829320k used, 241496k free, 106696k buffers Swap: 2048276k total, 0k used, 2048276k free, 503844k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ Command 4047 yoshiki 25 0 2780m 2.7g 2036 R 99.9 35.3 123:36.30 xhpl 4226 root 16 0 1048 1048 800 R 4.2 0.0 0:01.56 top 1 root 15 0 256 256 208 S 0.0 0.0 0:07.96 init 2 root RT 0 0 0 0 S 0.0 0.0 0:00.00 migration_CPU0 3 root RT 0 0 0 0 S 0.0 0.0 0:00.00 migration_CPU1 --------------------------------------------------------------------------- The xhpl process is using about 7GB memory. However, top command shows 2780m on VIRT and 2.7g on RES. So there is a turnup at 4GB. It is small problem since the process run well (performed 2.48GFLOPS). regards, Yoshiki Yamaguchi -- Yoshiki Yamaguchi SofTek Systems, Inc. yoshiki(a)softek.co.jp

2 2

RE: [suse-amd64] MSI K8D Master
by david.obrien＠amd.com 06 Aug '03

06 Aug '03

> so here is my configuration: > MSI K8D Master Version A9131KMS > BIOS Version is 1.00 What is the "V" number after 1.00 (as seen in the BIOS setup screen)? What is the date of your BIOS? -- David

3 3

MSI K8D Master
by Gregor Stößer 05 Aug '03

05 Aug '03

Hi there, I've seen that many of you are having trouble getting this board to work with SuSE 8.2 x86_64. It works (more or less) fine for me, so here is my configuration: MSI K8D Master Version A9131KMS BIOS Version is 1.00 2x AMD 240 CPU 1.4 GHz 6 x 1G RAM ECC + buffered Floppy Disk Drive 80GB IDE Hard Drive Maxtor 6Y080L0 IDE CDROM Tekram 390 SCSI Controller 2x U2W SCSI Hard Drive IBM DDRS-34560D U2W Hard Drive IBM DNES-309170W Enermax 400 Watt Power Supply The root file system is on the first scsi drive, boot loader is grub, residing in the mbr of the ide disk. Installation worked fine for me, I did not recognized major problems (in contrast to GinGin64 from Red Hat and Madrake 9.0 who both refused to boot with more than 4GB of RAM). So the only suggestion I have is that there is somethings wrong with the IDE chipset during installation (I don't really know !!). Feel free to contact if you have further questions. Good luck ! Gregor -- -*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*- Gregor Stößer email: gregor.stoesser(a)chemie.uni-karlsruhe.de Institut für Anorganische Chemie Universität Karlsruhe Tel: 0721/608 2988 Fax: 0721/608 4854 *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-

1 0

OpenPBS Queueing system in 8.2beta9
by Fred P. Arnold 02 Aug '03

02 Aug '03

Hello, Pardon my density, but when OpenPBS 2.3 from the rpms provided on the beta disks is installed on 8.2Beta9, and you attempt to activate it by 'chkconfig pbs_server on', the error insserv: Service pbsserv has to be enabled for service pbsserv insserv: exiting now! /sbin/insserv failed, exit code 1 is coughed up. The services have been added to /etc/services pbs 15001/tcp # pbs server (pbs_server) pbs_mom 15002/tcp # mom to/from server pbs_resmom 15003/tcp # mom resource management requests pbs_resmom 15003/udp # mom resource management requests pbs_sched 15004/tcp # scheduler pbsserv 15005/tcp This is different from our previous hand-PBS installation on AIX 4.2 and NPACI-ROCKS (which makes no mention of configs beyond the standard stock ones from the manul). Any suggestions, as it's probably something simple and SuSE related that I missed. Thank you in advance. -Fred Frederick P. Arnold, Jr. NUIT, Northwestern U. f-arnold(a)northwestern.edu

3 2

Install problem with 8.2 Beta 9 for AMD64
by Louie Athanasiadis 01 Aug '03

01 Aug '03

Hello, I am having trouble installing the beta. I have a 2 processor AMD Opteron 242 - 1.6Ghz MSI K8D Master (AMD 8131 Chip Set) - MS-9131(v1.X) SSI Mainboard It has 6GB of ram and a 40Gb IDE harddrive If has a removable Harddrive rack install (for using multiple OS's) and when the Window XP Professional 32 bit OS disk is installed it works fine. all images (disks 1 to 3) were downloaded and the MD5 sums generated and compared with the values on the website all were correct. The images were burned to CD's. When I install the SUSE Linux beta using all default settings (except time zone) it goes through the installation of disk 1 and reboots at the end of disk 1 without asking for disk's 2 or 3. After rebooting the system hangs with blank screen after flashing some text on the screen (which I have been unable to capture). If I boot with "Linux - Safe Settings" option it displays to these two lines and hangs Starting kswapd bigpage subsystem: allocated 0 bigpages (=0MB). Any Ideas how to get my $11K box running linux??? Regards Louie

2 1