Re: How to install Debian with AMD Opteron support?
On Fri, 2005-02-18 at 16:58 +, Rupert Heesom wrote: > As a newbie I want to try installing Debian on an dual-Opteron box that > I'm proposing to get. Read this document: https://alioth.debian.org/docman/view.php/30192/21/debian-amd64-howto.html Download this CD: http://debian-amd64.alioth.debian.org/install-images/sid-amd64-netinst.iso Have fun. -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Big filesystems.
On Tue, 2005-02-15 at 16:50 -0500, David Wood wrote: > On Tue, 15 Feb 2005, Jeffrey W. Baker wrote: > > That said, XFS is still your best choice if you've hit the hard limits > > in ext3. > > Ahh... _that_ said, it looks like (until they fix it) XFS is the best > choice for punishing your enemies with. :o > > I hate to say it, but this is not the only place I have heard Linux/XFS > horror stories. Of course I actually love experimenting, and there's > nothing wrong with a work in progress, just so long as it's labeled. > > I guess the moral of the story is that if you've got a big partition, I > hope you've got an even bigger backup tape. ;) Maybe the moral is you should use what everybody else uses. ext2/3 and to some extent Reiser are very well tested because practically everybody uses them. Major bugs in ext3 are readily apparent because it has millions of users. Major bugs in XFS are not found until one of sgi's 17 customers happens to trip over one. Worse, XFS is the sole user of lots of in-kernel code. ext3 uses lots of underlying code that is also used by other kernel pieces. The segmentation helps in rooting out problems. It's a many-eyes problem. There's a "Safe Operating Area" for most software. Probably if you went off and made a 4TB ext3 volume with the journal on a iSCSI device and then scattered it with 150 million files, you would run into a problem. That's unknown territory. But staying within a normal < 1TB volume with < 10 million files will almost certainly work without a problem. Once I worked at a company and we bought NetApp filers. NetApp sales engineers told us there was no such thing as fsck for WAFL, their filesystem. 6 months later we encountered something uniquely horrible known as a "wack". "Oh," they suddenly remembered, "there's no fsck unless..." We were not a mainstream user. They had never tested our workload. We switched to VxFS on EMC. 6 months later VxFS destroyed an extremely large filesystem. We were out of the mainstream. Never been tested. We switched to many small Linux servers with medium-sized ext2 partitions. No problems ever. End of fable :) -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Big filesystems.
On Tue, 2005-02-15 at 16:18 -0500, Adam Skutt wrote: > What kernel? What situation. Unfortunately, especially as of late, > there have been several known "gotcha" situations that will cause data > corruption, especially under 2.6 (4K stacks being one, very full > fileystems another). And the only common thread to all of these is XFS. You can say that maybe NFS is the problem, maybe this or that is the problem but the fact is that all the other filesystems survive in the environment and not XFS. Note that CIFS causes the same corruption that NFS does. Is Samba also broken? Note that XFS will corrupt a loop device on any kernel. I guess the loop device isn't "below" XFS in your world view? XFS is broken because it uses its own buffer management layer, imported to the tune of 1000s of lines of code from Irix. It maybe be in the kernel but it doesn't use kernel facilities and/or it uses them incorrectly. If you want to have a very large filesystem with lots of files *and* you are only going to use it in the "normal" fashion - directly attached storage not exported via NFS or CIFS - *and* you are only going to use RHEL or SuSE kernels (not kernel.org) then you will probably be fine. But you should realize what you are getting into. -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Big filesystems.
On Tue, 2005-02-15 at 15:43 -0500, Adam Skutt wrote: > Jeffrey W. Baker wrote: > > Unfortunately XFS also repeatedly swallowed a number of my volumes. I > > found it to be more unstable than any filesystem I have used (save > > VxFS). When using XFS, one must not read from the underlying device, or > > one risks corruption. > In Linux, doing this on *any* filesystem will potentially cause corruption. > > This is why e2dump/e2restore are *unsafe* and not to be used. > > Raw I/O operations bypass the buffer cache, so of course it'd be corrupted. The latter part of this sentence is not supported by the former, > If you're dicking with device access while a partition is mounted and > you lose data, you deserve what you get. This behavior has been a no-no > forever. Reading from the device never causes corruption of any other filesystem. You can dump an ext3 filesystem all day long. You'll get a corrupt dump, but you won't get a corrupt volume. > This leads one to believe that using XFS on LVM, > > md, or enbd would be somewhat risky. > I dunno what 'enbd' enbd is the enhanced network block device. > is, but XFS on LVM or md is perfectly safe. No, actually, it isn't. Look at the XFS mailing list. There's a number of reports of boinked volumes just in the last month. > They're > kernel drivers for starters, so they can coordinate block I/O > operations. They also sit below XFS in the I/O layer, XFS just sees > them as another partition. This does not explain why XFS is frequently reported to become corrupted when exported over NFS. > I dunno how you came across this conclusion at all. I came across this conclusion by losing numerous large filesystems in the course of only 6 month before abandoning XFS. -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Big filesystems.
On Tue, 2005-02-15 at 14:59 -0500, Adam Skutt wrote: > Kyle Rose wrote: > > > > What bothers me is file deletion time. Anyone have any clue why > > ReiserFS takes so long to delete files, and why the delete operation > > evidently blocks all other FS operations? It seems that ReiserFS > > should log the delete, and then have a kernel thread handling cleanup > > in the background in such a way that it doesn't cause other operations > > to block. > It's because ReiserV3 has to rebalance two b-trees: one for the > metadata, and one for the actual data itself. This is slow, especially > on large directories, I'd imagine. Surely you aren't implying that Reiser uses anything as pedestrian as a b-tree! Why, Reiser's tree format is so novel, so utterly perfect, that no human could have ever thought of it. I understand their patent applications is sailing through the approval process, greeted by nothing but disbelief and Hosannas. Right, so in my experience XFS destroys all other filesystems on metadata operations. XFS also formats and mounts the volume much more quickly than Reiser or ext2/3. XFS also supports very large volumes, very large files, and more directory entries than any of the others. I used to have a giant test tar archive that only XFS could have extracted in my expected lifetime. Unfortunately XFS also repeatedly swallowed a number of my volumes. I found it to be more unstable than any filesystem I have used (save VxFS). When using XFS, one must not read from the underlying device, or one risks corruption. This leads one to believe that using XFS on LVM, md, or enbd would be somewhat risky. fsck.xfs is sometimes at a loss to recover anything at all in these situations, even after running for days. That said, XFS is still your best choice if you've hit the hard limits in ext3. -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Big filesystems.
On Mon, 2005-02-14 at 14:49 -0500, Ben Russo wrote: > I have a Big disk array with 3 1.6TB RAID 5 LUNs and a global hot spare. > I want to bind them together into a single 4.8TB filesystem with LVM. > > I have an AMD Opteron Processor in the server to attach to the array. > I assume that I need a 64bit 2.6 series kernel yes? > Are there any special build options necessary, or do the regular 64-bit > 2.6 kernels allow such large filesystems? ext3 with the standard 4KB block will allow a 4TB filesystem. XFS is practically unlimited. 2.6 would clearly be better. 2.4 barely works on AMD64. -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
RE: Oracle won't cooperate
On Thu, 2005-02-10 at 16:48 -0800, Joseph Pelle wrote: > Did anyone get this (Oracle) working? Which Oracle download did you use? I > tried the linux x86-64, but I can't even get the runInstaller to work. You have to edit the runInstaller and associated programs before they will run. I was unable to get the Java installer to run. Note to Oracle: Java is not portable if you link in a bunch of 32-bit JNI garbage. The only way I was able to get this to work was to install Oracle on a RedHat system and transplant the installation onto a proper Debian system. -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
blown away by the new installer
I must say how impressed I am with the new Debian installer. I just grabbed the lastest amd64 ISO and installed on a machine with 2 CPUs, 8GB main memory, 2 IDE hard disks, three SCSI HBAs, a FibreChannel HBA, and two gigabit ethernet ports, *and* I installed the system onto a software RAID 1 root and boot without even the slightest hint of a problem. Absolutely unbelievable. I believe this is the coolest installer I've ever seen. SuSE "Enterprise" (snort) was unable to cope on the same box. Bravo! -jwb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Oracle won't cooperate
On Thu, 2005-01-06 at 22:46 -0500, Javier Kohen wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > > I fear I'm suggesting the obvious, but do you have the locales package > installed in the chroot? Yeah, I have the chroot configured for the same locales as the main system (and the system with the X server, too). I'm fairly certain the locale message is completely spurious and misleading. -jwb
Oracle won't cooperate
I tried to install Oracle 10g on a sid/amd64 machine just now. What a nightmare. I grabbed the java sdk from blackdown, installed the ia32-libs packages and a fullblown ia32 chroot as well. Still, Oracle just refuses to start. Their installer is a 32-bit launcher program which starts java which tries to call a bunch of native methods. I installed openmotif (wretch!) in both 64-bit and 32-bit lib trees, but to no beneficial effect. What a nightmare! Would it kill them to ship the installer for their 64-bit database in, oh, I dunno, a 64-bit binary?!?! The best part is the totally misleading error message: current locale is not supported in X11 Not! How about "calling 32-bit libs from 64-bit java from 32-bit binary on 64-bit system not supported"? Any successful Oracle installs out there? -jwb
Re: Firefox instability
On Tue, 2005-01-04 at 12:08 -0600, John Goerzen wrote: > On Tue, Jan 04, 2005 at 11:56:08AM -0600, John Goerzen wrote: > > I've only just tried FireFox on amd64 today. (Using Konqueror > > previously) > > OK, make that "Yesterday", oops. > > But in general, the symtoms I've seen are: > > 1. Random crashing > > 2. The firefox UI freezes for about 5-10 seconds, followed by either >normal behavior (60% of the time) or a crash (40% of the time) This may sound odd, but are you using any Motif, Lesstif, or OpenMotif applications? I've seen Motif apps screw up the clipboard handling so badly that FireFox will freeze for 5-10 seconds at a stretch trying to acquire the selection (for example, when you focus the URL bar). But I've never seen a crash. FWIW I use FireFox with XFWM and GNOME.
Re: Firefox instability
On Tue, 2005-01-04 at 11:28 -0600, John Goerzen wrote: > Hi, > > Firefox seems to be quite unstable on amd64. It's crashed on me many > times, though the 32-bit version seems fine on the same machine. Any > ideas? Is this just a general observation, or started recently? I haven't had the first problem with FireFox and I use it fairly heavily (as I suspect do most of the list members). Data-dependent system trouble (crash with sigill perhaps)? Odd plugins? JVM?
Re: [Fwd: Re: my new superdesktop...]
On Mon, 2004-12-20 at 16:12 -0800, Joaquin Menchaca wrote: > Where can one get a 64bit JVM? You can get the 1.4.2 software from java.blackdown.org, or the 1.5.0 bits from java.sun.com.
Re: xutils 4.3.0.dfsg.1-9 breaks xdm (bug #285699)
On Wed, 2004-12-15 at 14:33 +0100, Frederik Schueler wrote: > Hi, > > Maybe this is http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=285150 > hitting you? Yes, that fixes it. Thanks. -jwb
xutils 4.3.0.dfsg.1-9 breaks xdm (bug #285699)
I just dist-upgraded my amd64 wkstn, and xdm is broken. There seems to be something wrong with the sessreg program, and nobody can log in via xdm. I was able to workaround by disabling sessreg in /etc/X11/xdm/xdm.options Anyone else seeing this? If you haven't got this version yet you might want to hold off on upgrading. -jwb
Re: vmware and linuxthreads
On Thu, 2004-12-02 at 22:37 -0600, A. P. Kennedy wrote: > Here is mine: > > > ldd /usr/local/lib/vmware/bin/vmware > linux-gate.so.1 => (0x) > libdl.so.2 => /ia32/lib/tls/libdl.so.2 (0x55587000) > libm.so.6 => /ia32/lib/tls/libm.so.6 (0x5558b000) > libX11.so.6 => /emul/ia32-linux/usr/X11R6/lib/libX11.so.6 (0x555ae000) > libXext.so.6 => /emul/ia32-linux/usr/X11R6/lib/libXext.so.6 > (0x55675000) > libXi.so.6 => /emul/ia32-linux/usr/X11R6/lib/libXi.so.6 (0x55683000) > libgdk_pixbuf.so.2 => /emul/ia32-linux/usr/lib/libgdk_pixbuf.so.2 > (0x5568b000) > libglib-1.2.so.0 => /emul/ia32-linux/usr/lib/libglib-1.2.so.0 > (0x556a) > libgmodule-1.2.so.0 => /emul/ia32-linux/usr/lib/libgmodule-1.2.so.0 > (0x556c2000) > libgdk-1.2.so.0 => /emul/ia32-linux/usr/lib/libgdk-1.2.so.0 > (0x556c5000) > libgtk-1.2.so.0 => /emul/ia32-linux/usr/lib/libgtk-1.2.so.0 > (0x556fd000) > libpthread.so.0 => /ia32/lib/tls/libpthread.so.0 (0x55845000) > libc.so.6 => /ia32/lib/tls/libc.so.6 (0x55854000) > /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x5000) OK, thanks for confirming this. I will just leave this nugget of knowledge in the mailing list's archive for posterity: To run vmware 4.5.2 on Debian/amd64, your dynamic linker needs to use the libc from Debian/x86, rather than the libc in the ia32-libs package. -jwb
Re: vmware problem
On Fri, 2004-12-03 at 09:34 +0800, [EMAIL PROTECTED] wrote: > I remove the gcc-3.3 and install gcc-3.4,gcc-3.4-base with synaptic > ,but i don't remove gcc-3.3-base because if i remove it ,many other > pakages will be removed ,like apt... >From these errors, it looks like you are either running a 32-bit kernel, or your are trying to compile 32-bit modules. Are you trying to install vmware from within the chroot? That won't work. You need to install it from your normal 64-bit root, under a running 64-bit kernel. -jwb
Re: vmware and linuxthreads
On Thu, 2004-12-02 at 18:49 -0600, A. P. Kennedy wrote: > >>>>> "Jeffrey" == Jeffrey W Baker <[EMAIL PROTECTED]> writes: > > Jeffrey> On Wed, 2004-12-01 at 22:38 -0600, A. P. Kennedy wrote: > >> Wow, who said that. Anyway I'm using vmware just fine on AMD64. > >> Take a look a the mail archive for details. This has been > >> discussed in detail. The trick is gettting, and installing the > >> vmware-any-any-update. Make sure you use the latest version. Note > >> some strange error comes up when starting vmware just restart the > >> VM session and it works fine. > > Jeffrey> It seems you are running vmware under the 32-bit chroot. Is > Jeffrey> that correct? I'm trying to run it natively. > > Currently running natively under debian amd64 gcc-3.4. That is the > only reason I don't boot into 32 bit anymore. It can be done, and > works well. Really? I only though you were using a chroot because you said so yourself only a month ago: On Thu, 2004-11-04 18:03:35 -0600, A. P. Kennedy <[EMAIL PROTECTED]> wrote in message <[EMAIL PROTECTED]>: > kernel is 2.6.8.1 and all needed debs to support vmware are > installed in my chroot directory, and the correct lib directories are > placed in /etc/ld.so.conf So i guess you aren't really in the chroot, but you're using the libs installed there. Is that right? I have temporarily fixed my problem by installing the 32-bit libraries in a chroot and making sure they appear in /etc/ld.so.conf before the ia32-libs versions of the same libraries. So far this has allowed me to install Windows XP in vmware. Not sure if it's a good long-term solution, and if Debian/ia32 every switches to NPTL I'll be up a creek. APK, what's the output of ldd /usr/local/lib/vmware/bin/vmware? Here's mine: linux-gate.so.1 => (0x) libdl.so.2 => /var/chroot/sid-ia32/lib/tls/libdl.so.2 (0x55583000) libm.so.6 => /var/chroot/sid-ia32/lib/tls/libm.so.6 (0x55586000) libX11.so.6 => /emul/ia32-linux/usr/X11R6/lib/libX11.so.6 (0x555aa000) libXext.so.6 => /emul/ia32-linux/usr/X11R6/lib/libXext.so.6 (0x55671000) libXi.so.6 => /emul/ia32-linux/usr/X11R6/lib/libXi.so.6 (0x5567f000) libgdk_pixbuf.so.2 => not found libglib-1.2.so.0 => /emul/ia32-linux/usr/lib/libglib-1.2.so.0 (0x55687000) libgmodule-1.2.so.0 => /emul/ia32-linux/usr/lib/libgmodule-1.2.so.0 (0x556a8000) libgdk-1.2.so.0 => /emul/ia32-linux/usr/lib/libgdk-1.2.so.0 (0x556ac000) libgtk-1.2.so.0 => /emul/ia32-linux/usr/lib/libgtk-1.2.so.0 (0x556e4000) libpthread.so.0 => /var/chroot/sid-ia32/lib/tls/libpthread.so.0 (0x5582c000) libc.so.6 => /var/chroot/sid-ia32/lib/tls/libc.so.6 (0x5583b000) /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x5000) -jwb
Re: vmware and linuxthreads
On Wed, 2004-12-01 at 22:38 -0600, A. P. Kennedy wrote: > Wow, who said that. Anyway I'm using vmware just fine on AMD64. Take a > look a the mail archive for details. This has been discussed in > detail. The trick is gettting, and installing the > vmware-any-any-update. Make sure you use the latest version. Note some > strange error comes up when starting vmware just restart the VM > session and it works fine. It seems you are running vmware under the 32-bit chroot. Is that correct? I'm trying to run it natively. -jwb
Re: vmware problem
On Thu, 2004-12-02 at 08:45 -0600, A. P. Kennedy wrote: > > "vm" == <[EMAIL PROTECTED]> writes: > > vm> Hi, I just installed kernel-image-2.6.8-9-amd64-k8 and > vm> kernel-headers-2.6.8-9-amd64-k8 from Debian/unstable. I also have > vm> gcc 2.95.4 installed and made sure my /usr/bin/gcc was pointing > vm> to gcc-2.95.4 (and the same for g++). > > vm> and I download the vmware-any-any-update84.tar.gz > > vm> then I run the runme.pl in the vmware-any-any-update84 > > > Use gcc-3.4, and you need to compile a complete kernel and modules. > Make sure the new kernel is installed correctly and working. Then > run "runme.pl". I have never tried to compile kernel modules for > a packaged kernel, but suspect this will not work. It should work fine > then. Since I compile the kernel anyway this has never been an issue > for me. So the question is can one compile a kernel module without > actually having compiled the kernel? I don't know the answer to this > question. The trick to compiling these vendor modules is to install kernel-tree-2.6.9 or whatever the corresponding package is for your kernel. After that's installed you will be able to compile nvidia, vmware, and other 3rd-party kernel modules. But you still won't be able to run vmware, on account of nptl. -jwb
Re: x86_64 on sata raptor drive.
On Wed, 2004-12-01 at 21:50 +, CHRIS WAKEFIELD wrote: > Greetings all. > > I'm thinking of buying a 36 gig wd raptor 10,000 rpm blah, blah. > > Does anyone have any suggestions Re: installing x86_64 debian on this drive? > > Special options, drivers and such? Will the install kernel see the drive? I haven't had any problems with this drive (actually the 72GB version), but the real question is will the kernel see the SATA controller. Drives are not a problem normally, controllers are a pain. I'm using nforce3, had no trouble on the install. Check the K8 boards FAQ on alioth for more info. -jwb
vmware and linuxthreads
It seems vmware can't be used on Debian amd64 because glibc lacks linuxthreads support. Is it possible to build a glibc with nptl and linuxthreads at once? -jwb
Firefox plugin downloader pretends to install flash
The new Firefox has a cute behavior. If you click on that ANNOYING plugin downloader bar, it offers to download Flash, presents you with the EULA, then downloads and claims to successfully install the plugin. There's an upstream bug in Bugzilla on this behavior: https://bugzilla.mozilla.org/show_bug.cgi?id=265536 It's harmless, but is there any way to disable that annoying plugin downloader? It takes up a lot of screen space. -jwb
Re: Nvidia for amd64
On Wed, 2004-11-10 at 05:30 +0100, Harald Dunkel wrote: > I would like to contribute some updated nvidia-graphics-drivers > packages (1.0-6629) for amd64. This stuff is based upon > Randall's packages for i386 and the changes done by Markus > Benning a few months ago. I am completely in favor of Debian packages for this stuff, espcially since the 6111 driver did not install on Debian/amd64, but by the way I'll point out that the 6629 driver installs and runs perfectly* on my amd64 system with GeforceFX card. So it seems NVidia are making some progress, no surprise given their recent statement that 20% of their wkstn shipments are for Linux, and undoubtedly many of those are going in Opteron machines. -jwb * Actually I had to set CC=gcc-3.4 in the environment before installing.
1394 SBP2 still a crapshoot
First time I plugged in my firewire disk to my Althon64 PC, it loaded and mounted fine. I exported it via NFS and copied a lot of junk onto it. Since then, despite reboots and power cycles and a kernel upgrade, I cannot get the stupid disk to appear on the 1394 bus. Believe me, I've modprobed everything 8 ways to sunday. ieee1394: Error parsing configrom for node 0-01:1023 ieee1394: The root node is not cycle master capable; selecting a new root node and resetting... ieee1394: Error parsing configrom for node 0-00:1023 ieee1394: Node changed: 0-00:1023 -> 0-01:1023 This is so frustrating. _Years_ after 1394 storage became widespread we still have this completely useless 1394 stack. Argh. -jwb
Re: 64-bit kernel on sid / pure64?
On Tue, 2004-10-12 at 17:32, Kyle Rose wrote: > Stephen Waters <[EMAIL PROTECTED]> writes: > > > 2) Can 64-bit apps run on a 32-bit Xserver? > > Thanks to the magic of the client/server X protocol, all applications > that don't use the DRI should work fine no matter where the client is > running in relation to the server. This is mainly true, but I've had problems with the clipboard and Motif applications running on 32-bit machines with a local 64-bit X server. Namely the clipboard doesn't work in Nedit.
Re: recommendations for a barebone system
On Wed, 2004-10-13 at 03:36, Pablo Pita wrote: > El MiÃrcoles, 13 de Octubre de 2004 10:00, Eric Lacroix escribiÃ: > > I have this box, the sn85g4. I'm happy with its features. It runs quiet > > if you set the fans at very low speed, and don't get too hot (with > > Athlon64 3000+) (but not cool either). > > Alas, it happens a problem when it gets its stabilized temperature : the > > network interface drop out. Not working anymore. I checked on several > > forums, and I'm not alone. I've tried to cool it with an extra quiet > > fan, open the box ... but can't find what part of the mobo to cool. > > Only set the fan speed at average fix the problem (what I don't too > > because of the noise). > > For now, I run with an extra ethernet board, but it use my unique PCI > > slot. > > I have an SN85g4. The trick with the ethernet is you need to boot it with "noapic" on the kernel command line. It works perfectly that way. Otherwise the ethernet will stop working after a few seconds of heavy traffic. The only other problem I had, at the time, was the 1394 port showed up as eth0. I understand that's been fixed in the installer. If you have this machine I recommend you use the powernowd package. It runs your CPU down to 1GHz when it's not busy. My machine, which has been in continuous use for a few months now, is running with a CPU temp of 40C, fan speed 2200RPM. Even under heavy load, with the CPU at 2.2GHz, the CPU temp stabilizes at 52C. The cooler in this machine is efficient and quiet. Unfortunately I can't check the hard drive temperature because the SATA driver doesn't support SMART (yet). BTW I'm using the stock Debian kernel without problems. -jwb
Re: Don't see much difference between i386 vs amd64
On Wed, 2004-09-08 at 09:37, Mario Bertrand wrote: > Hi, > > I have successfully install debian with sid-amd64-monolithic.iso on my > second ide (+kernel 2.6.8-3) and I don't see much difference in the > performance vs i386 sarge (+kernel 2.6.8-1) on my first ide. It is > normal? Should I expect more? Someone told me that I should buy a new > hard drive with more cache (8M) to take benefit of faster cpu. > > hda - 2048k cache - 3923.96 BogoMIPS / i386 > hdb - 69k cache - 3932.16""/ amd64 BogoMIPS just tells you about the clock speed of your CPU, roughly speaking. As the name says, it's Bogus and Misleading. In my experience the 64-bit mode of the Opteron gains about 20% over the 32-bit mode, and of course you can address more memory per process. -jwb
Re: Success report: installing pure64 on a dell poweredge 1850
On Tue, 2004-09-07 at 14:10, Frederik Schueler wrote: > Hello, > > I had the chance to test the installation of pure64 on a dell poweredge > 1850 with "nocona" xeons, and I only can say: it works like a charm. > This system will presumably run sarge for amd64 as soon as it is > released. We have a Nocona Dell Precision Workstation that also boots the amd64 port. > TEST: Iterations/sec. : Old Index : New Index > : : Pentium 90* : AMD K6/233* > :--:-: > NUMERIC SORT: 923.36 : 23.68 : 7.78 > STRING SORT : 118.88 : 53.12 : 8.22 > BITFIELD: 3.4425e+08 : 59.05 : 12.33 > FP EMULATION: 125.88 : 60.40 : 13.94 > FOURIER : 12468 : 14.18 : 7.96 > ASSIGNMENT : 22.785 : 86.70 : 22.49 > IDEA: 2697.5 : 41.26 : 12.25 > HUFFMAN : 1619.6 : 44.91 : 14.34 > NEURAL NET : 16.017 : 25.73 : 10.82 > LU DECOMPOSITION: 507.28 : 26.28 : 18.98 > ==ORIGINAL BYTEMARK RESULTS== > INTEGER INDEX : 49.428 > FLOATING-POINT INDEX: 21.244 > Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler > 10.0 > ==LINUX DATA BELOW=== > C compiler : gcc-3.4 (unknown version) > libc: unknown version > MEMORY INDEX: 13.163 > INTEGER INDEX : 11.747 > FLOATING-POINT INDEX: 11.783 > Baseline (LINUX): AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38 > * Trademarks are property of their respective holder. Who can resist this? ;) On Athlon64 3400+ 1024K cache, gcc 3.4 TEST: Iterations/sec. : Old Index : New Index : : Pentium 90* : AMD K6/233* :--:-: NUMERIC SORT: 1685.8 : 43.23 : 14.20 STRING SORT : 197.92 : 88.44 : 13.69 BITFIELD: 4.627e+08 : 79.37 : 16.58 FP EMULATION: 162.4 : 77.93 : 17.98 FOURIER : 18897 : 21.49 : 12.07 ASSIGNMENT : 23.215 : 88.34 : 22.91 IDEA: 4043.1 : 61.84 : 18.36 HUFFMAN : 1673.3 : 46.40 : 14.82 NEURAL NET : 31.163 : 50.06 : 21.06 LU DECOMPOSITION: 1149.3 : 59.54 : 42.99 ==ORIGINAL BYTEMARK RESULTS== INTEGER INDEX : 66.894 FLOATING-POINT INDEX: 40.010 Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0 ==LINUX DATA BELOW=== CPU : AuthenticAMD AMD Athlon(tm) 64 Processor 3400+ 2194MHz L2 Cache: 1024 KB OS : Linux 2.6.7-5-amd64-k8-smp C compiler : gcc-3.4 libc: MEMORY INDEX: 17.324 INTEGER INDEX : 16.234 FLOATING-POINT INDEX: 22.191 Baseline (LINUX): AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
SIGFPE in motion
I was trying to get the motion package (detects motion in video capture) working on amd64. I grabbed it and unfortunately it was getting SIGFPE. I tried to debug it, but damned if I can't get gdb to produce a reasonable stack trace. The FPE might (MIGHT) be getting delivered somewhere in libjpeg. I can't really tell. Anyone else running into this with motion or another jpeg application? I tried rebuilding motion with various cflags but, no luck so far. -jwb
Re: acroread works now && /emul naming scheme?
On Mon, 2004-08-30 at 09:38, Ryan Lovett wrote: > On Mon, Aug 30, 2004 at 03:44:13PM +0200, Sebastian Steinlechner wrote: > > There, "evgeny" describes how to get acroread to work with amd64. The > > solution boils down to adding > > > > export XLOCALEDIR=/emul/ia32-linux/usr/X11R6/lib/X11/locale/ > > > > to the top of /usr/bin/acroread, as the locale directory already exists > > on pure64. > > Thanks for the tip. Is there an amd64 deb for acroread? I get the i386 > from Christian Marillat's site, though he doesn't yet have an amd64 > archive. I contacted him and he says he doesn't intend to start an AMD64 repository, but he will gladly add a link to one, the same way he links to the PPC repository. I have built most of his packages on AMD64 (save for mplayer, which is Hard), but I wouldn't want to host it for fear of the MPAA/DVD-CCA inquisition. If anyone wants to try to fix up the mplayer package, or host an AMD64 repository in Ukraine, drop me a mail. -jwb
Re: amd64 and video card experiences?
On Sun, 2004-08-22 at 16:38 +0300, Kyuu Eturautti wrote: > > On Sat, Aug 21, 2004 at 08:28:50PM -0600, Marcelo E. Magallon wrote: > >> So, if you want current-generation hardware you have no choice but > >> NVIDIA, which is available now and works. > > > > Not exactly. > > > > Current generation X (for example X.Org) seems to run fine on current > > generation ATI. [Well... it runs fine on 9600 and 9800 -- I'm only > > presuming it runs fine on X800.] > > > > It's the 3d acceleration which is not yet supported on amd64 for ATI. > > Two questions coming to my mind - first, to clarify, am I right in > assuming that if I don't want any 3D, I don't have to go toss my Radeon > away to run nothing more complex than a browser and a terminal in X on > AMD64? I have problems with my Radeon, even with plain 2D desktop graphics. On a Radeon 9700 Pro, there is noise of red dots all over the screen, at 1600x1200. If you use the ATI binary drivers, there is an option "TMDSCoherentMode", whose only effect seems to be to change the red dots to blue dots. Sigh. On another Radeon 7000-series card, which appears to have two DVI ports on the back, I cannot get the DVI output to work with the XFree86/Xorg drivers. ATI's drivers are required, and even then I can't get both DVI ports to work at once. > As for Nvidia, there's a lot of talk about Geforce series cards, but I've > too often found them to be unstable and quite hot (just imo, no flamebait > intended). So from Nvidia, I prefer the Quadro series. How's their > functionality, are there any experiences? I have a Quadro NVS at the office. It came with a Dell machine. With NVidia's drivers on Debian i386, it works fine (fast 2D and 3D). Even the quad head(yes FOUR outputs) works well. On another machine I have removed the fan from a GeForce 5200. Without using the 3D features, it still runs cold to the touch. Probably if you want a cheap card with *all* features supported in open source drivers, you should pick up an ATI Rage 128. The PPC people and others have really done a fine job with the driver. -jwb
Re: K8 Mainboards List
On Tue, 2004-08-17 at 08:27, Gordon Ball wrote: > I've used the 'intel8x0' driver with the Gigabyte GA-K8N Pro motherboard > onboard sound with reasonable success (with ALSA). > > This was in Gentoo system with a 2.6.7 kernel. > > Whether this applies to the other Gigabyte mainboards with the same onboard > sound (Realtek ALC658 Audio AC'97 Codec) I'm not sure, perhaps anyone with > these boards could confirm. I think intel8x0 will work with any nforce-based motherboard and most any AC'97 codec. I have a Shuttle board with nforce3 and Realtek ALC650 (rev 3). It works fine. -jwb
gdb broken on amd64 or omit-frame-pointer or ???
I was trying to debug a crasher in Firefox, but I keep getting these bogus stack traces out of gdb: > #210 0x0062776a2f656d6f in ?? () > #211 0x5f6e653d474e414c in ?? () > #212 0x382d4654552e5355 in ?? () > #213 0x373d53454e494c00 in ?? () > #214 0x3d4c564c48530031 in ?? () > #215 0x2f3d454d4f480031 in ?? () > #216 0x62776a2f656d6f68 in ?? () > #217 0x445f454d4f4e4700 in ?? () The first 48 levels of the stack are correct, but followed by several hundred levels of junk. I this problem caused by build flags in firefox package, or maybe some bogosity in gdb? Or some other thing? -jwb
