January 2004 - openSUSE AMD64 - openSUSE Mailing Lists

teTeX available for x86-64?
by Liaw, Andy 21 Jan '04

21 Jan '04

Hi, We have a dual Opteron box running SLES8. Does anyone know if there's teTeX for this? I could not find it in the CDs. TIA for any pointers. Best, Andy Andy Liaw, PhD Biometrics Research PO Box 2000, RY33-300 Merck Research Labs Rahway, NJ 07065 mailto:andy_liaw@merck.com <mailto:andy_liaw@merck.com> 732-594-0820 ------------------------------------------------------------------------------ Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD and in Japan as Banyu) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. ------------------------------------------------------------------------------

2 1

SUCCESS! SuSE9 x86_64 & 2.6.1-mm4 on Tyan 2885 & Sil3114
by Steven A. DuChene 21 Jan '04

21 Jan '04

This evening I was FINALLY successful at getting SuSE9 x86_64 fully installed and running on a Tyan S2885 (K8W) motherboard using the on-board Sil3114 SATA controller with two Seagate 72Gb drives hung off of them. I was able to install directly to the SATA drives without using any intermediate IDE or SCSI drives or controllers. The process was rather involved but now it is running a 2.6.1-mm4 kernel with the siimage driver that is part of the 2.6.1-mm4 kernel code tree. Unfortunately the closed source driver available from the ftp.tyan.com site for SuSE 9.0 x86_64 won't provide a bootable system but using the boot iso and the floppy image file from the tyan site gets you installed and to a shell prompt so you can pull down the 2.6.1 kernel source and Andrew Morton's mm4 patch for that 2.6.1 kernel. Then next is compiling a custom 2.6.1-mm4 kernel to get the system completely bootable. The closed source driver on Tyan's ftp site is compiled for a 2.4.21-102 kernel but the 9.0 X86_64 stuff released on the SuSE ftp sites (and mirrors) is built around 2.4.21-149 so once the system is installed the Tyan Sil6514.0 module won't actually load at a normal boot up. You can however use the boot.iso to boot and then stuck the floppy in prior to the kernel fully coming up. It will force load the module (something I couldn't figure out how to do in the context of the initrd part of a regular boot up) and then let you do the install and boot to a shell prompt afterwards to build a 2.6 kernel with the siimage community open source driver in it. One thing though, the closed source driver sees the controller and disks as SCSI devices (/dev/sda in my case) but the open source siimage driver sees them as IDE devices. So once you build and install your new 2.6.1-mm4 kernel prior to rebooting into it you have to change all the instances of sda (or whatever scsi disk it is seeing) in the fstab and lilo.conf file to the appropriate hd device (/dev/hde in our case). WARNING: DO NOT RUN the Setup_9.0 or SMP_Setup_9.0 outlined in step 3.4c of their directions without backing up your ORIGINAL lilo.conf FIRST! Very important as their scripts overwrite the original with a completely new fresh copy that is NOT based on the original created during the SuSE install process. Since their scripts and modules are built around a 2.4.21-102 kernel the lilo.conf file will not even work for a system that is 2.4.21-149 plus the root partition they specify in their lilo.conf file is probably not the same as what your install may be built around. -- Steven A. DuChene linux-clusters(a)mindspring.com sduchene(a)mindspring.com

2 1

Re: [suse-amd64] Solving the Tyan 2885 AGP performance issues
by Kevin_Gassiot＠veritasdgc.com 21 Jan '04

21 Jan '04

Nope, I haven't even tried to work with the RAID chip yet... I have been trying to get the graphics to work properly on an AMD64 platform since about last May, I want to take a few days away from beating my head against the wall, before I start banging against another wall :) Kevin Gassiot Advanced Systems Group Visualization Systems Support Veritas DGC 10300 Town Park Dr. Houston, Texas 77072 832-351-8978 kevin_gassiot(a)veritasdgc.com |---------+---------------------------> | | Andrew Halliwell| | | <ah(a)gnd.com> | | | Sent by: | | | ah(a)gnd.com | | | | | | | | | 01/21/2004 11:04| | | AM | | | | |---------+---------------------------> >---------------------------------------------------------------------------------------------------------------------| | | | To: suse-amd64(a)suse.com | | cc: | | Subject: Re: [suse-amd64] Solving the Tyan 2885 AGP performance issues | >---------------------------------------------------------------------------------------------------------------------| And verily, didst Kevin_Gassiot(a)veritasdgc.com announce to the hordes: > > > > > > The stock 102 kernel, 4499 driver, and stock BIOS will not work. You need > to update the BIOS, update the kernel, and update the Nvidia driver.... The nvidia driver on the site was still on 4499 when I read the original message, must've been hitting a webcache somewhere. It's there now. > I am running the S2885101k BIOS from Tyan, the 171-smp kernel from SuSE > with the mtrr.c replaced by the one supplied by AMD, and the 5331 driver > from Nvidia. You need all of these various bits to get the board to work > correctly. I have been working with these guys for some months now, trying > to get things to work, and as of Monday, I have a truly functioning system > :) You've not got the RAID on the silicon image on-board SATA working though, I'll wager... :) Just a pain that the kernel won't patch with those from that message. Obviously too old now to accept them. Just have to try to cheat the driver into working, by compiling a later 2.4.21- kernel and putting the build number in the makefile back to 102. -- Check the List-Unsubscribe header to unsubscribe For additional commands, email: suse-amd64-help(a)suse.com

1 0

Re: [suse-amd64] Solving the Tyan 2885 AGP performance issues
by Andrew Halliwell 21 Jan '04

21 Jan '04

And verily, didst Peter Rundberg announce to the hordes: > Hmmm. Quite... > Maybe you should try to apply the pathches that Mark Langsdorf suggested > on the list since your kernel does not have the fixes it needs to find > the AGP tunnel. > > The patches can be found here. > http://www.x86-64.org/lists/discuss/msg04476.html Looks like 102's not agp/mtrr-patch ready. Both failed. (Well, the agp one managed to patch one chunk, but I don't think one chunk of a patch is safe? the mtrr one failed with a malformed line 25) I wonder if I downloaded the 178 sources, and changed the makefile to 102... If it'd trick the Sil3114 driver into loading... That's the only thing keeping the 102 kernel here.

1 0

Problems with SuSE 9.0: new PC with Sata, (soft)raid, Graphics
by Andreas Huennebeck 21 Jan '04

21 Jan '04

Hi everybody, I'm planning to buy a new PC which should have a (soft)raid 1 disk system and a sensible 2D graphics card without fan. Due to its speed and the cool'n quiet option I got an offer about a PC with Asus K8VD, AMD64 3200+, two Maxtor Sata disks, a Plextor Premium CD-writer and an MSI GF4-MX440 graphics card. With this information I started to look for Linux support ( I want to stay with SuSE) and was quite shocked when I had to find out that this combination is very problematic: - Sata does not work at all, neither with the Via nor with the Promise controller on the motherboard. - The graphics card is not supported. In fact, searching on the SuSE hardware database for graphicscard with full or partial support on SuSE 9.0 AMD64 lists not one single card! This shurely is no satisfying state. Is this a problem due to binary drivers from the graphics card vendors? If yes, are there graphicscards (which you can buy now) which have an open source driver? So what should I do? - Use the standard 32Bit SuSE9, will this support the hardware listed above? - Use an EIDE softraid instead of Sata? This solution needs an additional EIDE controller as a softraid needs both channels and then there is no channel left for the CD-writer. - Use a Pentium P4 instead? Are there any Motherboards where the Sata controller is supported? Graphicscards should be no problem then, right? I can wait until end of march, then I need the new PC. Are there any chances that a new SuSE 9.1 distribution might be available which solves these problems? Thanks in advance Andreas -- Andreas Hünnebeck | email: ah(a)bruker.de

3 4

[suse-amd64] Problems with SuSE 9.0: new PC with Sata, (soft)raid, Graphics (fwd)
by Andrew Halliwell 21 Jan '04

21 Jan '04

Forwarded message: > Hi everybody, > > I'm planning to buy a new PC which should have a (soft)raid 1 disk system > and a sensible 2D graphics card without fan. Due to its speed and the cool'n > quiet option I got an offer about a PC with Asus K8VD, AMD64 3200+, two > Maxtor Sata disks, a Plextor Premium CD-writer and an MSI GF4-MX440 > graphics card. With this information I started to look for Linux support ( I > want to stay with SuSE) and was quite shocked when I had to find out that > this combination is very problematic: > > - Sata does not work at all, neither with the Via nor with the Promise controller > on the motherboard. I've looked at the 2.6.1 kernel and it's supported there. (non raid) If you go to tyan's site, there's a driver. It's binary only alas, so you'll be stuck with the kernel that came on the DVD. (that can do non-raid and RAID 1... But not RAID0 yet) > - The graphics card is not supported. I really doubt that. > In fact, searching on the SuSE hardware > database for graphicscard with full or partial support on SuSE 9.0 AMD64 lists > not one single card! That's probably the SuSE database that's behind, not the drivers. Linux doesn't use proprietory drivers for simple 2D cards. Only 3D acceleration requires those, and then only on some (like nVidia). This shurely is no satisfying state. > Is this a problem due to binary drivers from the graphics card vendors? If > yes, are there graphicscards (which you can buy now) which have an open > source driver? It might help if you said what the actual card WAS. Make, model number, etc. > So what should I do? > > - Use the standard 32Bit SuSE9, will this support the hardware listed above? Anything that works in 2D on X in 32bit should work on 64bit. (unless I'm mistaken)

2 1

Changing the stack limit for 32-bit programs
by Odon Farkas 21 Jan '04

21 Jan '04

Hi, I was trying to compile Gaussian 03 (mostly fortran, compiled with PGI) in 32-bit mode, (SuSE 9.0, x86_64), because it is currently faster then the 64-bit version. However, I ran into a stack overflow, because it seems that ld adjusts the stack limit to the old default 2MB. The current ld does not support the "--stack" flag. How can I change the stack limit then? Thanks a lot for your help in advance! Best wishes, Ödön Farkas

2 1

Re: [discuss] Re: [suse-amd64] Solving the Tyan 2885 AGP performance issues
by Steven A. DuChene 21 Jan '04

21 Jan '04

On Wed, Jan 21, 2004 at 05:54:31AM -0800, Jeffrey W. Baker wrote: > On Wed, 2004-01-21 at 02:06, Steven A. DuChene wrote: > > I don't believe this list permits attachments as no mtrr.patch was included > > with your E-mail. > > Must be a local problem, as I received the patches. -jwb Yes, but I am betting you are on the discuss(a)x86-64.org mailing list rather than the suse-amd64(a)suse.com mailing list... -- Steven A. DuChene linux-clusters(a)mindspring.com sduchene(a)mindspring.com

1 0

RE: [suse-amd64] Solving the Tyan 2885 AGP performance issues
by Alan Gray 21 Jan '04

21 Jan '04

Fantastic. > -----Original Message----- > From: mark.langsdorf(a)amd.com [mailto:mark.langsdorf@amd.com] > Sent: Tuesday, 20 January 2004 9:58 AM > To: suse-amd64(a)suse.com > Cc: discuss(a)x86-64.org > Subject: [suse-amd64] Solving the Tyan 2885 AGP performance issues > > > The 2P Opteron based Tyan 2885 K8W has some problems > with AGP/openGL graphics performance when using > nVidia graphics cards under Linux. AMD, Tyan, and > nVidia have been working together to develop fixes > for the problems. Four problems have been resolved > for the Tyan 2885 design. > > The first issue was the inabilty of the Linux kernel > to find the AMD-8151 AGP Tunnel. The AMD-8151 AGP > controller is not at the normal PCI address, and the > Linux kernel had to be modified to account for the > change. SUSE Linux kernels starting with > k_smp-2.4.21-139 include the change. The two > attached patches (tyan_agp.generic.patch and > tyan_agp.RHEL3.patch) fix the problem for most kernels > and for the Red Hat 2.4.21-4EL kernels, respectively. > > The second issue was a misallocation of MTRRs in the > Tyan BIOS that prevented the nVidia driver from > correctly caching the AGP aperture. Tyan fixed this > with the 2885101i.ROM (1.01i) BIOS, available at > ftp://ftp.tyan.com/bios_beta/2885101i.zip. > > The third issue involved the nVidia 10-4499 graphics > driver. nVidia has corrected the problems exposed > by the updated AGP and BIOS code and released the > 10-5332 driver to their website on January 20th. > > The final issue involved the Linux kernel MTRR code. > A patch that fixes the problem is attached to this > message (mtrr.patch). It has not been included in > any production kernels at this time. > > These four updates, taken together, allow the 2885 > to provide excellent 3D performance under Linux. > > AMD would like to thank all those who contributed to > defining, debugging, and verifying this problem. > > Mark Langsdorf > Team Lead, Linux Device Drivers > Advanced Micro Devices > mark.langsdorf(a)amd.com > > >

1 0

iommu=fullflush.... what's the performance hit?
by spike1＠ridcully.fsnet.co.uk 21 Jan '04

21 Jan '04

I've been looking for info regarding how much of a performance hit fullflush causes... Does anyone know, cos I've heard it can make quite a difference. Also, does anyone know if a more permanent fix to the problem that caused the kludge to be implemented in the first place is close, so the hit can be removed allowing speed and stability? Thanks.

1 0