March 2007 - openSUSE AMD64 - openSUSE Mailing Lists

IOMMU...PCI memory hole...MTRR...I give up.
by David W. Hankins 26 May '07

26 May '07

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

5 13

[opensuse-amd64] slow floppy after upgrade to opensuse 10.2
by Joe Georger 19 Apr '07

19 Apr '07

Hey all, I upgraded from Suse 9.3 to openSuse 10.2 (both x86_64) on my dual Opteron box and now the floppy drive is writing at about 1k/sec. It is mounted in the /etc/fstab with /dev/fd0 /media/floppy auto noauto,user,sync 0 0 Any ideas or help? Thanks, Joe --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

4 6

[opensuse-amd64] 4GB memory problem with OpenSuSe 10.2
by Peter Trifonov 11 Apr '07

11 Apr '07

Hello, I have got a PC based on GIGABYTE GA-965P-S3 motherboard (Intel P965 chipset, Core 2 Duo CPU) with 4GB memory. There are two problems with it: 1. OpenSuSe 10.2 x86-64 hangs during boot process with message "agpgart: detected an Intel 965G chipset". I managed to work around this problem by specifying mem=4096M boot option. 2. After the system has started, it reports that there is just 3.43Gb memory available. I was not able to find any options in BIOS related to memory hole remapping. Anyway, Windows XP x64 sees 4094 Megabytes. Is there any way to allow Linux to see all 4Gb of memory? With best regards, P. Trifonov --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

4 9

[opensuse-amd64] distribution of virtual address space
by Detlef Grittner 28 Mar '07

28 Mar '07

Hello, I have a question about the distribution of the virtual address space between 32 and 64 bit apps. This would affect the x86_64 kernel on AMD64 and the newer Intel processors, if I am correct. Assumed I have about 8 GByte on board and load 64 bit apps first, then they will go into the first 4 GByte of the virtual address space meaning that I can't load any 32 bit apps, when only the upper 4 GByte are left free? I'm asking on this discussion list, although I think there must be documentation, but I didn't find anything. Most likely I don't know what I need to look for exactly. Regards, Detlef --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

2 1

[opensuse-amd64] Cross-compiling SuSE kernel fails
by Frank Steiner 28 Mar '07

28 Mar '07

Hi, I'm trying to compile a set of i586 kernel rpms on a x86_64 host running SuSE 10.1(x86_64). I'm using the kernel-source-2.6.16.27-0.9.i586.rpm for the sources and compile them with "make ARCH=i386..." Works file for the kernels/modules, but generating the stuff for my kernel-source.rpm I call make ARCH=i386 O=/root/tmp/kernel/usr/src/linux-2.6.16-6suse-bio-obj/i386/default scripts which results in e.g. gcc -Wp,-MD,scripts/basic/.docproc.d -Iscripts/basic -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -o scripts/basic/docproc /root/tmp/kernel/usr/src/linux-2.6.16-6suse-bio/scripts/basic/docproc.c and I end up with "ELF 64-bit LSB executable, AMD x86-64" files: ...i386/default/scripts/basic/docproc: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for GNU/Linux 2.6.4, dynamically linked (uses shared libs), for GNU/Linux 2.6.4, not stripped So I get a kernel-source.i586.rpm that I can't install on a i586 host due to these binaries. All other stuff like the modules or the kernel are compiled with "gcc -m32", but that -m32 is missing here. When I call make HOSTCC="gcc -m32" scripts ... I get 32bit binaries, but I'm not sure if that's the right way to do it. Is there any reason why "make scripts" ignores the ARCH parameter? Is there a better fix than the "HOSTCC=.." call? cu, Frank -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. * --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

2 2

[opensuse-amd64] Re: [opensuse] hwinfo can't be rebuilt at 64bit Linux
by Patrick Shanahan 28 Mar '07

28 Mar '07

* Zhao Shujing <gldyszh(a)gmail.com> [03-27-07 22:40]: > I try to rebuild hwinfo at RHEL for 64-bit. But there are some errors: [...] > Anyone have met this status? Someone in one of the RH lists may have, but why rebuild? If you are running openSUSE, hwinfo is provided, hwinfo-12.29-3, on the install iso. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/gallery2 OpenSUSE Linux http://en.opensuse.org/ --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

1 0

[Fwd: Re: [opensuse-amd64] distribution of virtual address space]
by Dave Howorth 27 Mar '07

27 Mar '07

On Fri, 2007-03-23 at 12:16 +0100, Detlef Grittner wrote: > I have a question about the distribution of the virtual address space > between 32 and 64 bit apps. This would affect the x86_64 kernel on AMD64 > and the newer Intel processors, if I am correct. > > Assumed I have about 8 GByte on board and load 64 bit apps first, then > they will go into the first 4 GByte of the virtual address space meaning > that I can't load any 32 bit apps, when only the upper 4 GByte are left > free? You don't need to worry. I think you're confusing virtual address space with physical address space. You could google for those terms, and memory mapping. Cheers, Dave --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

3 2

[opensuse-amd64] SuSe 10.0/64 4Gb memory problem
by James D. Parra 21 Mar '07

21 Mar '07

Hello, Installed 4GB of physical memory, however only 3GB appear; # cat /proc/meminfo MemTotal: 3342628 kB I'm running the 64 bit kernel (Suse 10.0); 2.6.13-15-smp #1 SMP Tue Sep 13 14:56:15 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux ...with two CPU's; # cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 37 model name : AMD Opteron(tm) Processor 246 stepping : 1 cpu MHz : 2009.270 <snip> processor : 1 vendor_id : AuthenticAMD cpu family : 15 model : 37 model name : AMD Opteron(tm) Processor 246 stepping : 1 cpu MHz : 2009.270 cache size : 1024 KB <snip> Any ideas on how to remedy this? Many thanks, ~James --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

8 11

RE: [opensuse-amd64] SuSe 10.0/64 4Gb memory problem
by James D. Parra 19 Mar '07

19 Mar '07

"Joachim Deguara" <joachim.deguara(a)amd.com> wrote: > James D. Parra wrote: > > Hello, > > > > Installed 4GB of physical memory, however only 3GB appear; > > > > # cat /proc/meminfo > > MemTotal: 3342628 kB > > It sounds like you need to go into the BIOS and ask for "Memwhole > remapping" and set it to "Software". This happens because PCI takes the > address space from 3-4GB (staying under the 32bit boundary) and maps > over where memory may be so we want the BIOS to map that memory elsewhere. (disclaimer: I work for AMD in the CPU architecture team) Remapping in the BIOS is possible with all current AMD parts: -- First off, some BIOSes have bugs, so test it out before just using it! -- If this is a pre-rev-E part (C0 or CG), then you can perform "Software" remapping, but it slows down memory access a bit on Opteron systems. -- If this is a Rev E or later part, then you can perform "Hardware" remapping as well, which has no performance cost. I'd recommend using the "Hardware" method if you have a Rev E or later Opteron part. ~~~~~~~ Thank you for your responses. I used 'Hardware' for the Memwhole remapping. What anomalies might I expect if using 'Hardware' is not the right choice? Again, many thanks. ~James --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-amd64+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-amd64+help(a)opensuse.org

1 0

[opensuse-amd64] Some problems after upgrading to 10.2
by Constantine 'Gus' Fantanas 18 Mar '07

18 Mar '07

Hl all. Finally, I bit the bullet and upgraded to 10.2 (64-bit). I was expecting I would have to reinstall a few packages, such as MPlayer and the nvidia drivers, so I don't count them as problems. The problems I am seeing are two: (1) For some reason, rpm seems to consume over 99% of the CPU time. I have to kill it manually. (2) If I invoke yast from inside KDE as a normal user, yast does not seem to recognize the root password, but if I su to the root account from the command line, the very same password is recognized (as it should). Any ideas? TIA. -- 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

3 3