Re: nvidia driver installation problem
hi yannick, i got a detailed output from dmesg, i've supported rivafb in kernel 2.6.13 that i compiled, i had no idea that rivafb and nvidia conflicts. now i'll compile kernel again without rivafb support. lsmod ouptu didn't show rivafb and when i tried modprobe rivafb it was loaded. this means it is dynamic loadable module, isn't it? the thing that i didn't understand is it enough for rivafb support that this conflict happpens? i mean, neither i load this module or not, this conflict happens cause i gave rivafb support for kernel. thanks. Azer 18.09.2005 tarihinde Yannick - Debian/Linux <[EMAIL PROTECTED]> yazmış: > Hi, > > I can't help you with this particular problem but i maybe can give you > some hope : > i've tried to build the nvidia driver in debian way with 2.6.13 from > kernel.org but i failed. The good news is i installed 7676 with the > nvidia-installer (nvidia way) after removed the nvidia debian packages : > i'm using it currently and as an exemple bzflag and ennemy-territory > works very well here... But as i tried a lot of stuffs to get it work i > can't provide you a process to follow. > > Regards, > Yannick > > Azer Demir a écrit : > > >hi everybody, > > > >yesterday i compiled 2.6.13 kernel, cause i hope to see my tv card > >detected automatically. i did compile with make-kpkg, and used > >2.6.11-9-amd64-k8 kernel's config file. there was no problem during > >compile and after it. when i booted the system with 2.6.13 kernel, but > >kernel can't load nvidia module. i removed old nvidia-glx, > >nvidia-kernel-common packages, and prepapared for new nvidia driver > >installation. > > > >first, i run module-assistant and > >it made nvidia-kernel-2.6.13.20050917 package. after this, i got > >nvidia-graphics-drivers source package, and made the binary packages. > >but when i rebooted the system i saw that kernel > >still can't load nvidia module. now i want to give outputs of some > >commands. > > > >`dpkg -l | grep nvidia` > >- > >ii nvidia-glx1.0.7174-3 NVIDIA > >binary XFree86 4.x driver > >ii nvidia-kernel-2.6.13.200509171.0.7174-3+r1NVIDIA binary > >kernel module for Linux 2.6.13 > >ii nvidia-kernel-common 1.0.7174-1NVIDIA binary > >kernel module common files > >ii nvidia-kernel-source 1.0.7174-3 NVIDIA binary > >kernel module source > > > > > >`locate nvidia.ko` > > > >/lib/modules/2.6.13.20050917/nvidia/nvidia.ko > >/usr/src/modules/nvidia-kernel/debian/nvidia-kernel-2.6.13.20050917/lib/modules/2.6.13.20050917/nvidia/nvidia.ko > >/usr/src/modules/nvidia-kernel/debian/nvidia-kernel-2.6.8-11-amd64-generic/lib/modules/2.6.8-11-amd64-generic/nvidia/nvidia.ko > > > > > >the binary packages that i made by nvidia-graphics-drivers > >--- > >nvidia-glx_1.0.7174-3_amd64.deb > >nvidia-glx-dev_1.0.7174-3_amd64.deb > >nvidia-glx-ia32_1.0.7174-3_amd64.deb > >nnvidia-kernel-source_1.0.7174-3_amd64.deb > > > >at first, i suspected nvidia-kernel-common packet's different version > >number from others, i use testing, and i looked with apt-show-versions, > >nvidia-kernel-common is uptodate for testing, there is different > >versions in unstable and experimental. > > > >a question come on my mind that aren't these packages for nvidia driver > >enough, or did i install a package that is not necessary? > > > >a nvidia.ko module that i used with 2.6.8-11-amd64-generic kernel remain > >in the system as it can see from `locate nvidia.ko` output, can problem > >be a conflict of two 'nvidia.ko' module? i mean, didn't old nvidia > >driver files clean completely? > > > >if i didn't remember wrong, after nvidia driver installation, nvidia > >creates a devicce file(the name may be nvidia, isn't it?). i saw an > >error "no such device", after i can't login by gdm. and when i try to > >modprobe nvidia, i get the same error. > > > ># modprobe nvidia > >FATAL: Error inserting nvidia > >(/lib/modules/2.6.13.20050917/nvidia/nvidia.ko): No such device > > > > > >the last possibility i think that the device nvidia creates is missing. > > > >$ ls /dev > >cdromptyc7 ptyq9 ptyub ptyydtty22 ttyb1 ttyp3 ttyt1 ttyx3 > >cdrw ptyc8 ptyqa ptyuc ptyyetty23 ttyb2 ttyp4 ttyt2 ttyx4 > >console ptyc9 ptyqb ptyud ptyyftty24 ttyb3 ttyp5 ttyt3 ttyx5 > >core ptyca ptyqc ptyue ptyz0tty25 ttyb4 ttyp6 ttyt4 ttyx6 > >fb0 ptycb ptyqd ptyuf ptyz1tty26 ttyb5 ttyp7 ttyt5 ttyx7 > >fb1 ptycc ptyqe ptyv0 ptyz2tty27 ttyb6 ttyp8 ttyt6 ttyx8 > >fd ptycd ptyqf ptyv1 ptyz3tty28 ttyb7 ttyp9 ttyt7 ttyx9 > >fd0 ptyce ptyr0 ptyv2 ptyz4tty29 ttyb8 ttypa ttyt8 ttyxa > >full ptycf ptyr1 ptyv3 ptyz5tty3 ttyb9 ttypb ttyt9 ttyxb > >hdd ptyd0 ptyr2 ptyv4 ptyz6tty30 ttyba ttypc tt
Re: Debian Sid amd64 on a Compaq v2310us (status report)
On 9/20/05, Ken Bloom <[EMAIL PROTECTED]> wrote: I have previously posted to vox-tech@lists.lugod.org about attempts toinstall Linux on my Turion64 based Compaq Presario v2310us laptop. I'mposting this as a howto to both vox-tech, and also debian-amd64, so that there should be an internet record of how to get a successful systemworking.Ken, thanks for your post. From the specs of your laptop, it's fairly similar to my Acer Aspire 5024, (which probably explains why the acer wireless driver works ;) I have one open problem at the moment: I can't get the battery status to show. Did you have to change anything to make that work, or are you using a preconfigured kernel ? TimT.
Re: unreal tournament 2004
Probably this isn't helpful, but I didn't have to do anything special to run ut2004 - just ran the installer, it detected the amd64 system, installed the 64bit binaries, and it just ran. Can you get any useful error/logging output? -- Debianista!
Debian Sid amd64 on a Compaq v2310us (status report)
I have previously posted to vox-tech@lists.lugod.org about attempts to install Linux on my Turion64 based Compaq Presario v2310us laptop. I'm posting this as a howto to both vox-tech, and also debian-amd64, so that there should be an internet record of how to get a successful system working. In the beginning I tried installing Debian Sid for amd64, but gave up quickly because I didn't know of wireless drivers for the laptop. I didn't even keep that installed long enough to bother writing down what worked and what didn't. That problem could have been resolved way back then, but I tried a few more distributions before making Debian Sid work again. I tried Kubuntu hoary for i386, and the results are documented at http://lists.lugod.org/mailinglists/archives/vox-tech/2005-09/msg3.html and at http://lists.lugod.org/mailinglists/archives/vox-tech/2005-09/msg00017.html I would bet you're probably going to try to read these. I'm going to reference some things in those posts, but only for the sake of contradicting things I said in those posts. After finding that wireless driver, I reinstalled to Kubuntu breezy for amd64. That turned out to be a complete disaster, owing in part to some bugs still outstanding on Debian Sid (they're fixable though and irrelevant to most people), but also owing to a complete and total melt down of all of my network drivers, for reasons passing understanding. So I gave up and installed Debian Sid for AMD64 again. *** Executive Summary *** The wireless card needs ndiswrapper to run. The windows driver is available from acer. You can either get accelerated video with the fglrx driver but no resume support, or you can get unaccelerated video with the ati driver, but with resume support. I chose the latter. You need kernel 2.6.11 or later to get DMA working. I don't know anything about the memory card reader, nor do I have firewire peripherals. It seems that the former probably doesn't work, but the latter probably does. *** Specs *** *** Specs *** The specs for this laptop are as follows (copied from http://reviews.designtechnica.com/review2907_specs14544.html, with slight modifications to add slightly more techinal information) Processor:AMD Turion 64 processor ml-28+ with PowerNow! 1.60GHz Chipset: ATI Shipping operating system:Windows XP Professional for 32-bit with Service Pack 2 Memory:512MB DDR SDRAM (2 x 256MB) at 333MHz; maximum memory 2048MB DDR SDRAM Graphics card:ATI MOBILITY RADEON 200M with 128MB DDR (dedicated) Hard drive:80GB Enhanced IDE Primary CD/DVD drive:DVD/CD-RW combo drive Display:14.0" WXGA High-Definition BrightView Widescreen (1280 x 768) display Wireless connectivity:54g Integrated 802.11b/g wireless LAN with 25HSM/SpeedBooster support (based on a Broadcom chipset) Digital media:6-in-1 integrated Digital Media Reader for Secure Digital cards, MultiMediaCard, Memory Stick, Memory Stick Pro, SmartMedia or xD-Picture Cards (Texas Instruments) External ports:3 Universal Serial Bus (USB) 2.0, 1 notebook expansion port, 1 IEEE 1394 (FireWire) Communications:Integrated 10/100Base-T Ethernet LAN (RJ-45 connector), high speed 56K modem (RJ-11 connector) Dimensions:13.15" (L) x 9.1" (W) x 1.29" (H) (min) or 1.53" (H) (max) Weight:5.38 lb. Sound:Altec Lansing *** Debian Install *** I installed Debian Sarge, using a netinstall CD over wired ethernet. The install used Kernel 2.6.8 as the install kernel. As DMA for the ATI chipset is not supported until kernel 2.6.11, I had another working computer download kernel 2.6.12 from Debian sid amd64, so that I could scp it over and install it as soon as possible after the initial install is done. After installing the 2.6.12 kernel, I upgraded to Debian Sid, and installed my favorite packages. *** Video *** You can use the ATI driver, or the non-free fglrx driver. The ATI driver is unaccelerated, but supports suspend/resume. The fglrx driver is accelerated, but does not support suspend/resume. I chose the former. To specify that you want unaccelerated video, you must use the option Option "noaccel" in Section "Device". Becuase my cursor looked ugly after that, I also used Option "swcursor" You may be able to get away without swcursor. Try it. The worst that can happen is aesthetic problems. My xorg.conf is attached. For some reason, I can't use the scroll stuff, and can't get the typing-disables-trackpad behavior. I could probably reconfigure that, but I would have to use Option "shmconfig" "on" in the Section "InputDevice" corresponding to the synaptics touchpad. When I turned that on, things started to crash, so I turned it off. I might have even had that working on Debian and lost it in a dist-upgrade. I'm not sure. If you have a solution, let me know. You can ignore the other mutterings on vox-tech about synaptics not working. I know I've gotten it to work fine on at least one of these installs. *** Video Part
Re: Dupal Opteron on Sarge
Paul , thanks for the corrections below (Hypertransport), I mistook the word ;-o. I also agree with you that the cooling SHOULD be same for 246-252 , but my experience shows me that its not. The Zalman most certainly qualifies for all CPUs here ;-) (its one of the few around) , but its not necessarily small ;-) Best regards Nils Valentin Tokyo / Japan http://www.be-known-online.com >> Another concern is the heat those CPU beasts produce (90W each). You may >> want to find a cooler solution which doesnt grill you or the CPU and >> also >> doesnt make you death. (I am sitting next to such a beast right now ;-) >> >> Most cooling systems go up to the 246 or 248. Finding one for the 252 >> might be difficult. I had a hard time finding a cooler which would do >> his >> job for the 248 while still allowing me to hear my own voice ;-). > > IIRC the 252 has the same thermal rating (89W) as the 246, so the cooling > system should be the same. Even the dual-core chips are only 95W. Most > Opteron cooling systems should be able to cope with any of these. The > Zalman > coolers are very quiet and more than capable of coping with these CPUs. > > If you're bothered about heat/noise you should probably go for the HE > cpus. > These are more expensive, but only generate about half the heat (55W). A > 246HE runs exactly the same speed as a regular 246. > > You should also look at the Athlon64 X2 CPUs. For many workloads these are > just as good as a dual socket board. Obviously with a dual socket board > you > have the option of building a quad-core machine. > >> Note that 2xCPU systems are not twice as fast as single CPU systems. The >> SMP setup does have a bit of a overhead on your OS, so expect something >> like 0,8x the speed of a single CPU system. > > amd64 systems scale pretty well up to [at least] 4 CPUs. In my experience > much > better than Intel systems. I've seen CPU/memory intensive workloads scale > linearly. Obviously if your workload is IO bound throwing more CPUs at it > probably won't help at all. > >> What the 2xCPU system gives you however, is the ability to handle heavy >> load. It can handle obviously more requests than a single CPU system. It >> will also take advantage of the hyperthreading bus *communication bus >> between the CPUs. > > Hypertransport, not Hyperthreading. > >> This means that CPU1 can *borrow Memory from CPU2 if it is required for >> an >> application. > > It's worth noting that even remote memory (ie. attached to the other CPU) > is > still closer (lower latency) than system memory on many Intel systems. > > Paul > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact > [EMAIL PROTECTED] > > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Re: soundcard not detected by alsaconf
I have the Chaintech VNF4 Ultra VE. Its a nforce4 motherboard. I think that it has a realtek chip but I know that I can use the nvidia or Realtek drivers on the windows side. I am convinced that its a compatibility issue with this kernel but, if anyone know of any alternatives, I'm open to suggestions. I'm still also interested in a kernel install tutorial. thanks
Re: unreal tournament 2004
im running debian-amd64 sid also nvidia gefore 6600gt pcie nvidia drivers 1.0-7676 (previous versions dont like my card) k 2.6.12-1-amd64-k8 3d acceleration is working glxgears gives 3000+fps tuxracers works nicely at around 500fps gltron also with around 1000fps. needless to say though i havent paid $2500 for 1000fps gltron :( Dean If we can't figure out what's happening here, try asking on the ut2004 mailing list: http://icculus.org/lgfaq/#ut2k4ml Or, if you have an nvidia card, try the nvidia linux forum: http://www.nvnews.net/vbulletin/forumdisplay.php?f=14 For what it's worth, I have ut2004 working fine, so there's not some inherent problem with debian-amd64 and ut2004. * debian-amd64 sid * linux 2.6.13 * nvidia geforce 5900xt * nvidia drivers version 1.0-7676 -Corey -- WWW: http://dean.bong.com.au LAN: http://www.bong.com.au EMAIL: [EMAIL PROTECTED] or [EMAIL PROTECTED] ICQ: 16867613 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: unreal tournament 2004
i wish i could get into the game to have this bug! does ut2k4 access the cdrom? as the 2.6.12 kernel has a kernel oops when i try to mount the cdrom (other than that it works better than 2.6.8) any thoughts? is there a 2.6.12-2 or 2.6.13 deb around? Dean Alexander Nagel wrote: Hi, i cannot help you with your current problem. But if you can play check following bug: https://bugzilla.icculus.org/show_bug.cgi?id=2054 greets Alex -- WWW: http://dean.bong.com.au LAN: http://www.bong.com.au EMAIL: [EMAIL PROTECTED] or [EMAIL PROTECTED] ICQ: 16867613 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: tcmplex
On Tue, 20 Sep 2005 09:23, William Brown wrote: > As well, I tried searching debian archives sarge, sid, marillat, and > found nothing for tcmplex, or even mplex. > > Can anyone tell me where I can find tcmplex, or mplex for sid, in > sourced or precompiled binary? mplex is in mjpegtools, which you should find on marillat (and source on SF). Despite it's name, it has a range of mpeg, mpeg2 and mjpeg tools > Thanks. Hope this helps. Later Lee Begg pgpsgG9KDLXT7.pgp Description: PGP signature
tcmplex
I think I finally have some good parameters to the transcode utility to create some good DVD mpeg2 format videos. However, once I use transcode to transcode and split into video/audio files, I need to re-merge them back into one file. What I'm looking for is tcmplex. What I've found on the net with google tells me that this is the tool I need to merge my audio and video back into one file before burning it to disc with growisofs. I see that marillat debian sources are back up, at ftp.nerim.net, but I took alook there and don't see any tcmplex package or source, nor do I see it as a part of the transcode source archive. avimerge is there, but I need something to merge my mpeg output, not my avi files. As well, I tried searching debian archives sarge, sid, marillat, and found nothing for tcmplex, or even mplex. Can anyone tell me where I can find tcmplex, or mplex for sid, in sourced or precompiled binary? Thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: unreal tournament 2004
Dean Hamstead wrote: > i did a clean install then untarbz2'd the 3355 patch and then cp -R * > all its files into the /usr/local/games/ut2004 directory > > then in Xorg i jumped into a shell, went to > /usr/local/games/ut2004/System and ran ut2004-bin-linux-amd64 > > splash starts, screen goes black. computer reboots. most odd That shouldn't happen. Even if your ut2004 installation is screwed up I wouldn't expect it to be able to crash your machine. What video card do you have, and what driver version? If we can't figure out what's happening here, try asking on the ut2004 mailing list: http://icculus.org/lgfaq/#ut2k4ml Or, if you have an nvidia card, try the nvidia linux forum: http://www.nvnews.net/vbulletin/forumdisplay.php?f=14 For what it's worth, I have ut2004 working fine, so there's not some inherent problem with debian-amd64 and ut2004. * debian-amd64 sid * linux 2.6.13 * nvidia geforce 5900xt * nvidia drivers version 1.0-7676 -Corey -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
wake-on-lan with Abit KV-80
Hello, all. I've been having trouble getting wake-on-lan (WOL) working with some Abit KV-80 cluster nodes and was wondering if someone on one of these lists had any thoughts. (Maybe useful: sarge 2.6.8-11-amd64-k8). So far, I have only been able to get WOL working (with standard 'magic packets') from a fully powered off state, ie. the power to the board had been killed by turning off the power supply. I have NOT been able to get it to work after powering down the board from a "poweroff" or "shutdown" signal. I've also tried various combinations of the -H/-P/-h options with "shutdown". Does anyone have any idea why this might be the case? I have tried every possible permutation of BIOS and hardware settings to no avail. If WOL works from a completely powered off state, I can only assume that there is something peculiar about how "poweroff" is "powering down" the machine, ie. it's not fully powering down the machine in a way that mimics what happens when the power supply is cut. I guess this isn't a complete surprise. Can someone fill me in on what exactly is the state that the machine ends up in after a "poweroff", and why WOL might not work in that state? Is there a way to coax the machine into a powered off state where it WOULD be receptive to a WOL? Does this have something to do with ACPI? I'm not running acpid. The last thing the console spits out after a poweroff command is "acpi_power_off called", but I'm not certain how to interpret that. As you can imagine, Abit has not been of any help yet. Any help at all would be much appreciated. jamie. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dupal Opteron on Sarge
Yes, I did say natively. AMD really works overtime, and if I get a Athon64, and Athlon XP of the same clock speed will run the programs at the same speed if you could somehow remove the extra goodies from the Athlon64, and ensure that the entire system was equal. However, you are generally going to get better performance from an Athlon64, and I still reccommend a dual core Opteron, b/c in many ways it's like getting two Opterons in one package. God knows the bus speeds can handle it. And since the amd64 chips can natively run in 32 bit mode, you don't really have to worry about whether your programs are 64 bit or not, though you should by all means keep on the lookout for amd64 programs specifically for you, for that alleged 10% increase. I can't confirm that, nor can I deny, though I'm more inclined to believe it.
Re: Dupal Opteron on Sarge
I would guess compared to Intel's first line of "x86" 64bit CPU's, the Intanium's which had a 32 bit "emulation" mode which took serious performance hits. I haven't seen much on what their (Intel) new line of x86_em64 chips are like now though. Jacob Bresciani "Passwords are like bubble gum, strongest when fresh, should never be used by groups and create a sticky mess when left laying around" -anon On Sep 19, 2005, at 10:38 AM, Paul Brook wrote: They can natively run 32 bit stuff with NO speed penalities No speed penalty compared to what? Amd64 chips do run 32-bit code "natively", and are competitive as 32-bit CPUs in their own right. However you usually get a speed increase from recompiling as a 64-bit app. In 64-bit mode you have more registers and 64-bit arithmetic, which allow the compiler can generate more efficient code. From what I've seen 64-bit apps tend to be ~10% faster then the 32-bit equivalent. Paul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: hdparm not o.k. ? (update)
Am Montag, 19. September 2005 19:51 schrieb Jean-Luc Coulon (f5ibh): > Le 19.09.2005 19:04:20, Hans a écrit : > > Hi folks, > > I tried to improve my settings of the harddrive. These are the > > original > > settings: > > > > /dev/hda: > > multcount= 16 (on) > > IO_support = 1 (32-bit) > > unmaskirq= 0 (off) > > using_dma= 0 (off) > > keepsettings = 0 (off) > > readonly = 0 (off) > > readahead= 256 (on) > > geometry = 16383/255/63, sectors = 195371568, start = 0 > > > > You see, dma is off. When trying to set this to on ( hdparm -d 1 > > /dev/hda), I > > get this error: > > > > /dev/hda: > > setting using_dma to 1 (on) > > HDIO_SET_DMA failed: Operation not permitted > > using_dma= 0 (off) > > [ ... ] > > It seems that you have booted your system with ide-generic loaded > instead of the module related to your chipset. > > In your /etc/modules, you should have something like that: > > via82cxxx > ide-generic I suppose, the reason is the kernel itself. I made it with built in ide-generic, as I do not have an initrd.img. So I have to build a new kernel, with moduralized ide-generic. Thanks for the clue, when I have news, I will give you the info. Now for the build. Best regards Hans > I use the via module to use the dma. ide-generic is needed for the > cdrom to work. The modules should placed in this order. > > If you use an initrd, you should have the requested module loaded in it. > > > Could it be a bug ??? Any clue ?? > > > > Best regards > > > > Hans > > > > > > Look at this, please. > > Jean-Luc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: soundcard not detected by alsaconf
On Sun, Sep 18, 2005 at 12:27:08AM -0400, Tony Andrews wrote: > I've been trying all day to get the sound up and running, but no dice. I > have the chaintech vnf4 ultra which has onboard nvidia sound. I have the > snd-intel8x0 driver module and modprobe gives no error. I tried commenting > out the oss modules but I'm not to sure that I did a thorough job(im a newb) > Here's my lsmod output: Which chipset do you have? You may need a newer kernel (like 2.6.12) to support a fairly recent chipset's sound chip even though it is compatible with snd-intel8x0. If the driver doesn't know it should be able to work your chip's pci id, then it won't try. Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libc6 problem
On Sun, Sep 18, 2005 at 03:24:30PM -0700, lordSauron wrote: > Just a question, but does ia## indicate an Intel Itanium processor? ia32 = intel architecture 32bit aka i386/x86 ia64 = intel architecture 64bit aka itanium (not x86_64 which is amd64 architerture) Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Realtek ALC850 8-channel audio CODEC
On Sat, Sep 17, 2005 at 09:45:24PM +1000, Dean Hamstead wrote: > ok huray video works > on to sound > > this board has an Realtek ALC850 8-channel audio CODEC which in lspci is > CK804 AC97 audio codec. the realtek site claims this is supported with > snd-intel8x0 > > alsa-conf shows up nothing, modprobing it works but the card isnt detected > > any ideas? this kernel is a tad old - although its straight out of > 'unstable' If adding snd-intel8x0 to /etc/modules doesn't find the chip on boot, then you may need a newer kernel/newer alsa driver to support the new chip. Many chips are supported in 2.6.11+ that were not in 2.6.8 for example. Are you running the current kernel for unstable? (That would be 2.6.12). Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: hdparm not o.k. ?
Am Montag, 19. September 2005 19:51 schrieb Jean-Luc Coulon (f5ibh): > Le 19.09.2005 19:04:20, Hans a écrit : > > Hi folks, > > I tried to improve my settings of the harddrive. These are the > > original > > settings: > > > > /dev/hda: > > multcount= 16 (on) > > IO_support = 1 (32-bit) > > unmaskirq= 0 (off) > > using_dma= 0 (off) > > keepsettings = 0 (off) > > readonly = 0 (off) > > readahead= 256 (on) > > geometry = 16383/255/63, sectors = 195371568, start = 0 > > > > You see, dma is off. When trying to set this to on ( hdparm -d 1 > > /dev/hda), I > > get this error: > > > > /dev/hda: > > setting using_dma to 1 (on) > > HDIO_SET_DMA failed: Operation not permitted > > using_dma= 0 (off) > > [ ... ] > > It seems that you have booted your system with ide-generic loaded > instead of the module related to your chipset. Yes, you are right ! In my /etc/modules is the entry "ide-generic" and not the via82cxxx. This one is beeing loaded by the kernel later. I will fix it. Maybe you are right, the ide-generic first loaded could disturb the via82cxxx-module. Good idea ! I will test it. Here is my /etc/modules ide-cd ide-disk ide-generic r8169 psmouse freq_table powernow-k8 cpufreq_userspace ide-scsi nvidia v4l2-common usbnet irport usb-storage firmware_class cpufreq_ondemand cpufreq_conservative Maybe this helps. Hans > > In your /etc/modules, you should have something like that: > > via82cxxx > ide-generic > > I use the via module to use the dma. ide-generic is needed for the > cdrom to work. The modules should placed in this order. > > If you use an initrd, you should have the requested module loaded in it. > > > Could it be a bug ??? Any clue ?? > > > > Best regards > > > > Hans > > > > > > Look at this, please. > > Jean-Luc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dupal Opteron on Sarge
> They can natively run 32 bit stuff with NO speed penalities No speed penalty compared to what? Amd64 chips do run 32-bit code "natively", and are competitive as 32-bit CPUs in their own right. However you usually get a speed increase from recompiling as a 64-bit app. In 64-bit mode you have more registers and 64-bit arithmetic, which allow the compiler can generate more efficient code. From what I've seen 64-bit apps tend to be ~10% faster then the 32-bit equivalent. Paul
Re: hdparm not o.k. ?
Le 19.09.2005 19:04:20, Hans a écrit : Hi folks, I tried to improve my settings of the harddrive. These are the original settings: /dev/hda: multcount= 16 (on) IO_support = 1 (32-bit) unmaskirq= 0 (off) using_dma= 0 (off) keepsettings = 0 (off) readonly = 0 (off) readahead= 256 (on) geometry = 16383/255/63, sectors = 195371568, start = 0 You see, dma is off. When trying to set this to on ( hdparm -d 1 /dev/hda), I get this error: /dev/hda: setting using_dma to 1 (on) HDIO_SET_DMA failed: Operation not permitted using_dma= 0 (off) [ ... ] It seems that you have booted your system with ide-generic loaded instead of the module related to your chipset. In your /etc/modules, you should have something like that: via82cxxx ide-generic I use the via module to use the dma. ide-generic is needed for the cdrom to work. The modules should placed in this order. If you use an initrd, you should have the requested module loaded in it. Could it be a bug ??? Any clue ?? Best regards Hans Look at this, please. Jean-Luc pgpNBPgcl5zmM.pgp Description: PGP signature
Re: Dupal Opteron on Sarge
On Sat, Sep 17, 2005 at 06:49:55PM -0700, lordSauron wrote: > oh... as for performance increases... > > pentium 4s use a 21 stage pipeline or something like that... so they > take approximately 21 clock cycles to get anything done. AMD uses > about 7 stages (or something in that neighbourhood) so if you divide > 2.8 by 21 and 2.0 (my Athlon64) by 7, you get a really interesting > breakdown. You'll certainly find a HUGE increase in performance, > ESPECIALLY if you go dual-core. As for dual-core and 64bit kernels > and the like... there was a few. you'd be looking at the > amd64-generic-SMP kernels (I think) the amd64 is for 64bit (x86_64) > and the generic is b/c I don't know if opterons are considered a k8 > thing (I'd have to go check) and the SMP is to give you dual-core > action, so you can use the withering power now avaliable to you. It's > soo cool... someday I'll get one... until then I just > annoy everyone with wordy descriptions of 'em. > > and for those of you who're mentally screaming at my crude way of > benchmarking the two cpus, it's only for a ballpark comparison. by no > means is that ENTIRELY accurate, and should NOT be treated as such. It's more like 31 on the current P4s and about 13 or 14 on the opteron. Most sane designs for x86 seem to use about 12 to 14 stage pipeline (athlon, opteron, pentium-m, etc). Good balance for clock speed vs cost of a branch miss. I believe the current high end powerpcs also run about that length of pipeline. The real problem with too long a pipeline is the cost in branch prediction misses and if there isn't enough independant instructions to keep the pipeline busy working. Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Pingus
On Sun, Sep 18, 2005 at 10:47:54PM -0500, Jaime Ochoa Malagón wrote: > I have this error... > > Preparing to replace pingus 0.6.0-8 (using .../pingus_0.6.0-8.1_amd64.deb) > ... > Unpacking replacement pingus ... > dpkg: error processing > /var/cache/apt/archives/pingus_0.6.0-8.1_amd64.deb(--unpack): > trying to overwrite `/usr/share/locale/locale.alias', which is also in > package locales > > Seems to be only amd64 and seems not good... See bugs.debian.org/327718 Kurt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: amd64 bootsplash howto
That is surely cleaner, but with this package, it seems you don't have the possibility to put a picture in the background of your consoles tty1-6, contrary to what you can do with bootsplash :( On 9/19/05, Rafael Rodríguez <[EMAIL PROTECTED]> wrote: > apt-get install -t experimental splashy > > El Jueves, 15 de Septiembre de 2005 15:00, Andrei Mikhailovsky escribió: > > Hello debian users, > > > > I am wondering if anyone managed to get graphical bootsplash to work > > with amd64. > > > > I've followed the howto, but the compilation of bootsplash package gives > > me the following error: > > > > > > gcc -Os -Wall -I/usr/include/freetype2 -c -o mng.o mng.c > > mng.c: In function 'copyline': > > mng.c:231: error: invalid lvalue in increment > > mng.c:282: error: invalid lvalue in increment > > make[1]: *** [mng.o] Error 1 > > make[1]: Leaving directory > > `/usr/local/src/bootsplash/bootsplash-3.1/Utilities' > > make: *** [build-stamp] Error 2 > > debuild: fatal error at line 765: > > > > > > The rest of the packages seems to compile and install fine, including > > kernel patches and sysv scripts. > > > > I've also read that there is splashy package for debian that takes care > > of the graphical boot process. There is an experimental version for > > debian i386 and some other platforms. However, i couldn't find splashy > > for amd64. Has anyone played with it on amd64. > > > > Many thanks for all the help > > > > Andrei > >
Re: hdparm not o.k. ?
Am Montag, 19. September 2005 19:38 schrieb Jacob Bresciani:
> The only other thing that jumps to mind is permissions, I assume
> you're running as root or through sudo?
Yes, I am running as root. Directly as root, not sudo.
Hans
>
> Jacob Bresciani
>
>
> "Passwords are like bubble gum, strongest when fresh, should never be
> used by groups and create a sticky mess when left laying around"
>
> -anon
>
> On Sep 19, 2005, at 10:19 AM, Hans wrote:
> > Am Montag, 19. September 2005 19:18 schrieben Sie:
> >> do you have the proper chipset driver for your HD controller or
> >> motherboard loaded? if not then DMA won't work because it won't know
> >> where/how to turn it on.
> >
> > Yes, I have. Chipset is via82. And the kernel module is loaded !
> >
> > Hans
> >
> >> Knoppix loads a kernel which probably has your MB /HD controller
> >> loaded so at least you know it will work.
> >>
> >> Jacob Bresciani
> >>
> >> "Passwords are like bubble gum, strongest when fresh, should never be
> >> used by groups and create a sticky mess when left laying around"
> >>
> >> -anon
> >>
> >> On Sep 19, 2005, at 10:04 AM, Hans wrote:
> >>> Hi folks,
> >>> I tried to improve my settings of the harddrive. These are the
> >>> original
> >>> settings:
> >>>
> >>> /dev/hda:
> >>> multcount= 16 (on)
> >>> IO_support = 1 (32-bit)
> >>> unmaskirq= 0 (off)
> >>> using_dma= 0 (off)
> >>> keepsettings = 0 (off)
> >>> readonly = 0 (off)
> >>> readahead= 256 (on)
> >>> geometry = 16383/255/63, sectors = 195371568, start = 0
> >>>
> >>> You see, dma is off. When trying to set this to on ( hdparm -d 1 /
> >>> dev/hda), I
> >>> get this error:
> >>>
> >>> /dev/hda:
> >>> setting using_dma to 1 (on)
> >>> HDIO_SET_DMA failed: Operation not permitted
> >>> using_dma= 0 (off)
> >>>
> >>> could there be a bug in hdparm ? When running knoppix in 32-Bit-
> >>> mode it works
> >>> fine.
> >>>
> >>> These are the infos from /dev/hda:
> >>>
> >>> hdparm -i /dev/hda
> >>>
> >>> /dev/hda:
> >>>
> >>> Model=ST9100823A, FwRev=3.01, SerialNo=3LG02H6A
> >>> Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.
> >>> 5% }
> >>> RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
> >>> BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16
> >>> CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=195371568
> >>> IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
> >>> PIO modes: pio0 pio1 pio2 pio3 pio4
> >>> DMA modes: mdma0 mdma1 mdma2
> >>> UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
> >>> AdvancedPM=yes: unknown setting WriteCache=enabled
> >>> Drive conforms to: ATA/ATAPI-6 T13 1410D revision 2:
> >>>
> >>> * signifies the current active mode
> >>>
> >>> Could it be a bug ??? Any clue ??
> >>>
> >>> Best regards
> >>>
> >>> Hans
> >>>
> >>>
> >>> Look at this, please.
> >>>
> >>>
> >>> --
> >>> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> >>> with a subject of "unsubscribe". Trouble? Contact
> >>> [EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: hdparm not o.k. ?
The only other thing that jumps to mind is permissions, I assume
you're running as root or through sudo?
Jacob Bresciani
"Passwords are like bubble gum, strongest when fresh, should never be
used by groups and create a sticky mess when left laying around"
-anon
On Sep 19, 2005, at 10:19 AM, Hans wrote:
Am Montag, 19. September 2005 19:18 schrieben Sie:
do you have the proper chipset driver for your HD controller or
motherboard loaded? if not then DMA won't work because it won't know
where/how to turn it on.
Yes, I have. Chipset is via82. And the kernel module is loaded !
Hans
Knoppix loads a kernel which probably has your MB /HD controller
loaded so at least you know it will work.
Jacob Bresciani
"Passwords are like bubble gum, strongest when fresh, should never be
used by groups and create a sticky mess when left laying around"
-anon
On Sep 19, 2005, at 10:04 AM, Hans wrote:
Hi folks,
I tried to improve my settings of the harddrive. These are the
original
settings:
/dev/hda:
multcount= 16 (on)
IO_support = 1 (32-bit)
unmaskirq= 0 (off)
using_dma= 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead= 256 (on)
geometry = 16383/255/63, sectors = 195371568, start = 0
You see, dma is off. When trying to set this to on ( hdparm -d 1 /
dev/hda), I
get this error:
/dev/hda:
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma= 0 (off)
could there be a bug in hdparm ? When running knoppix in 32-Bit-
mode it works
fine.
These are the infos from /dev/hda:
hdparm -i /dev/hda
/dev/hda:
Model=ST9100823A, FwRev=3.01, SerialNo=3LG02H6A
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.
5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=195371568
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: unknown setting WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 2:
* signifies the current active mode
Could it be a bug ??? Any clue ??
Best regards
Hans
Look at this, please.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: hdparm not o.k. ?
do you have the proper chipset driver for your HD controller or
motherboard loaded? if not then DMA won't work because it won't know
where/how to turn it on.
Knoppix loads a kernel which probably has your MB /HD controller
loaded so at least you know it will work.
Jacob Bresciani
"Passwords are like bubble gum, strongest when fresh, should never be
used by groups and create a sticky mess when left laying around"
-anon
On Sep 19, 2005, at 10:04 AM, Hans wrote:
Hi folks,
I tried to improve my settings of the harddrive. These are the
original
settings:
/dev/hda:
multcount= 16 (on)
IO_support = 1 (32-bit)
unmaskirq= 0 (off)
using_dma= 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead= 256 (on)
geometry = 16383/255/63, sectors = 195371568, start = 0
You see, dma is off. When trying to set this to on ( hdparm -d 1 /
dev/hda), I
get this error:
/dev/hda:
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma= 0 (off)
could there be a bug in hdparm ? When running knoppix in 32-Bit-
mode it works
fine.
These are the infos from /dev/hda:
hdparm -i /dev/hda
/dev/hda:
Model=ST9100823A, FwRev=3.01, SerialNo=3LG02H6A
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=195371568
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: unknown setting WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 2:
* signifies the current active mode
Could it be a bug ??? Any clue ??
Best regards
Hans
Look at this, please.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: unreal tournament 2004
Hi, i cannot help you with your current problem. But if you can play check following bug: https://bugzilla.icculus.org/show_bug.cgi?id=2054 greets Alex -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dupal Opteron on Sarge
well first of all, the raw power increase is going to be worthwhile. I'd suggest a nice 2xCPU server from that pcsforeveryone.com site or whatever it was called (they had some good deals) and customise for one CPU that's dual core - you can add another CPU later. The dual cores are sorta like getting two CPUs in many ways. Also, ALL AMD64 chips are FULLY backwards compat with ALL 32 bit operations. They can natively run 32 bit stuff with NO speed penalities - though it is nice to get the extra 32 bits when you find a 64bit program. I gotta run now, seeya later.
hdparm not o.k. ?
Hi folks,
I tried to improve my settings of the harddrive. These are the original
settings:
/dev/hda:
multcount= 16 (on)
IO_support = 1 (32-bit)
unmaskirq= 0 (off)
using_dma= 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead= 256 (on)
geometry = 16383/255/63, sectors = 195371568, start = 0
You see, dma is off. When trying to set this to on ( hdparm -d 1 /dev/hda), I
get this error:
/dev/hda:
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma= 0 (off)
could there be a bug in hdparm ? When running knoppix in 32-Bit-mode it works
fine.
These are the infos from /dev/hda:
hdparm -i /dev/hda
/dev/hda:
Model=ST9100823A, FwRev=3.01, SerialNo=3LG02H6A
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=195371568
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: unknown setting WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 2:
* signifies the current active mode
Could it be a bug ??? Any clue ??
Best regards
Hans
Look at this, please.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: dvd+rw-tools and dvdrecord problem / solution
On 9/17/05, Jean-Luc Coulon (f5ibh) <[EMAIL PROTECTED]> wrote: > Le 17.09.2005 04:04:00, Will Dyson a écrit: > Q. What are the dvd+rw-tools for? > A. As implied/already mentionned - to master the DVD media, both +RW/R > and -R[W]. . > > And in the package description: > Description: DVD+-RW/R tools > This package includes a format utility for DVD+-RW media, a frontend > to mkisofs for burning DVD+-RW/R and some associated tools. > > I'm a bit puzzled... After looking at the dvd+rw-format source code (http://fy.chalmers.se/~appro/linux/DVD+RW/tools/dvd+rw-format.cpp), it seems that I was wrong about dvd+rw-format not being useful at all for the -RW disc variety. It can perform blanking of DVD-RW (and DVD-RAM) media. -- Will Dyson
Re: Dupal Opteron on Sarge
> Another concern is the heat those CPU beasts produce (90W each). You may > want to find a cooler solution which doesnt grill you or the CPU and also > doesnt make you death. (I am sitting next to such a beast right now ;-) > > Most cooling systems go up to the 246 or 248. Finding one for the 252 > might be difficult. I had a hard time finding a cooler which would do his > job for the 248 while still allowing me to hear my own voice ;-). IIRC the 252 has the same thermal rating (89W) as the 246, so the cooling system should be the same. Even the dual-core chips are only 95W. Most Opteron cooling systems should be able to cope with any of these. The Zalman coolers are very quiet and more than capable of coping with these CPUs. If you're bothered about heat/noise you should probably go for the HE cpus. These are more expensive, but only generate about half the heat (55W). A 246HE runs exactly the same speed as a regular 246. You should also look at the Athlon64 X2 CPUs. For many workloads these are just as good as a dual socket board. Obviously with a dual socket board you have the option of building a quad-core machine. > Note that 2xCPU systems are not twice as fast as single CPU systems. The > SMP setup does have a bit of a overhead on your OS, so expect something > like 0,8x the speed of a single CPU system. amd64 systems scale pretty well up to [at least] 4 CPUs. In my experience much better than Intel systems. I've seen CPU/memory intensive workloads scale linearly. Obviously if your workload is IO bound throwing more CPUs at it probably won't help at all. > What the 2xCPU system gives you however, is the ability to handle heavy > load. It can handle obviously more requests than a single CPU system. It > will also take advantage of the hyperthreading bus *communication bus > between the CPUs. Hypertransport, not Hyperthreading. > This means that CPU1 can *borrow Memory from CPU2 if it is required for an > application. It's worth noting that even remote memory (ie. attached to the other CPU) is still closer (lower latency) than system memory on many Intel systems. Paul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: dvd+rw-tools and dvdrecord problem / solution
On Sat, Sep 17, 2005 at 07:52:46PM +, Jean-Luc Coulon (f5ibh) wrote: > If it doesnt work why can we read the following in the docs: > DVD+RW/+R/-R[W] for Linux > > and > > Q. What are the dvd+rw-tools for? > A. As implied/already mentionned - to master the DVD media, both +RW/R >and -R[W]. . > > And in the package description: > Description: DVD+-RW/R tools > This package includes a format utility for DVD+-RW media, a frontend > to mkisofs for burning DVD+-RW/R and some associated tools. Maybe the package description is wrong and it should say: This package includes a format utility for DVD+RW media, a frontend to mkisofs for burning DVD+-RW/R and some associated tools. Of course the docs for dvd+rw-tools say that ONLY dvd+rw needs formating, and growisofs actually auto detects if it hasn't been done and does it (in a few seconds). I imagine it probably would auto blank or overwrite a dvd-rw when asked to (with -Z), just as it does with dvd+rw. Otherwise perhaps the dvd+rw-format -blank=full would do the right thing to a dvd-rw. Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: burn cd/dvd: permission /dev/sg0 denied
> > but it tries to open hda, not hdc... > > I have no ideas... > > > > Can you help me? > > Well you could simply try: > cdrecord -dev ATAPI:/dev/hdc -toc (list toc of a cd) or -atip (list info > about a blank CDR type). > > If that works, you probably have what you need. > > Make sure ide-scsi is NOT loaded, and that ide-cd IS loaded. > > If it does DVD's use growisofs -[Z for new, M for append] /dev/hdc > [mkisofs options like -J -R] [files to burn] Hi Lennart, thanks for your reply. I remember it didn't solve (I have tried it in the past). However I have solved upgrading to kernel 2.6.12 + udev + hal. Sincerely, I don't know if it was kernel-debian 2.6.11 bug... Giulio -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: burn cd/dvd: permission /dev/sg0 denied
On Sat, Sep 17, 2005 at 08:39:28AM +0200, antongiulio05 wrote: > I have setted my system as: > > /usr/share/doc/cdrecord/README.ATAPI.setup > > under kernel 2.6.* section, but it doesn't work. > > I have launched: > > # cdrecord dev=ATA: -scanbus > Cdrecord-Clone 2.01.01a01 (x86_64-unknown-linux-gnu) Copyright (C) 1995-2004 > Joerg Schilling > NOTE: this version of cdrecord is an inofficial (modified) release of cdrecord > and thus may have bugs that are not present in the original version. > Please send bug reports and support requests to <[EMAIL PROTECTED]>. > The original author should not be bothered with problems of this > version. > > cdrecord: Warning: Running on Linux-2.6.11-9-amd64-k8 > cdrecord: There are unsettled issues with Linux-2.5 and newer. > cdrecord: If you have unexpected problems, please try Linux-2.4 or Solaris. > scsidev: 'ATA:' > devname: 'ATA' > scsibus: -1 target: -1 lun: -1 > Warning: Using badly designed ATAPI via /dev/hd* interface. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > Error trying to open /dev/hda exclusively (Device or resource busy)... > retrying in 1 second. > cdrecord: Device or resource busy. Cannot open '/dev/hda'. Cannot open SCSI > driver. > cdrecord: For possible targets try 'cdrecord -scanbus'. > cdrecord: For possible transport specifiers try 'cdrecord dev=help'. > cdrecord: > cdrecord: For more information, install the cdrtools-doc > cdrecord: package and read /usr/share/doc/cdrecord/README.ATAPI.setup . > > but it tries to open hda, not hdc... > I have no ideas... > > Can you help me? Well you could simply try: cdrecord -dev ATAPI:/dev/hdc -toc (list toc of a cd) or -atip (list info about a blank CDR type). If that works, you probably have what you need. Make sure ide-scsi is NOT loaded, and that ide-cd IS loaded. If it does DVD's use growisofs -[Z for new, M for append] /dev/hdc [mkisofs options like -J -R] [files to burn] Lennart Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
evolution and spell checking is now broken
Hi I am wondering if anyone experienced similar problems with Evolution and spell checking after a recent upgrade. I've noticed that aspell package has been updated and since then the spell checking is gone in Evolution. I've double checked that the gnome-spell and aspell (including dictionaries) is installed. All the packages seems to be in place, however spell checking fails to work. This seems to be the case for Abiword as well, the spelling icon can no longer be clicked, which seems to indicate a general gnome brokenness of spell checking :-( Anyone has any ideas on how to fix it? Andrei -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: unreal tournament 2004
Am Montag, 19. September 2005 14:50 schrieb Dean Hamstead: > i did a clean install then untarbz2'd the 3355 patch and then cp -R * > all its files into the /usr/local/games/ut2004 directory > > then in Xorg i jumped into a shell, went to > /usr/local/games/ut2004/System and ran ut2004-bin-linux-amd64 > > splash starts, screen goes black. computer reboots. most odd > > Dean > > > Hello ! > > Did you install the latest patch ? It is called > > "ut2004-lnxpatch3339.tar.bz2". > > > > Additionally you have to start the ut2004-amd64. Please check your > > start-script under /usr/local/bin. Maybe you have to edit. By default the > > 32-bin-version is beeing started. But you have to start with > > ut2004-bin-linux-amd64 ! > > > > Best regards > > > > Hans > > -- > WWW: http://dean.bong.com.au LAN: http://www.bong.com.au > EMAIL: [EMAIL PROTECTED] or [EMAIL PROTECTED] > ICQ: 16867613 That is weired. I just installed it as you did. After installation I went to /usr/local/games , and in this directory I did a chgrp -R games ut2004/ As I am in this group, I have success to all the files. Check your membership: you should be member of these groups: audio, video, games, users, dip, dialout, cdrom, plugdev Good luck Hans -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: unreal tournament 2004
i did a clean install then untarbz2'd the 3355 patch and then cp -R * all its files into the /usr/local/games/ut2004 directory then in Xorg i jumped into a shell, went to /usr/local/games/ut2004/System and ran ut2004-bin-linux-amd64 splash starts, screen goes black. computer reboots. most odd Dean Hello ! Did you install the latest patch ? It is called "ut2004-lnxpatch3339.tar.bz2". Additionally you have to start the ut2004-amd64. Please check your start-script under /usr/local/bin. Maybe you have to edit. By default the 32-bin-version is beeing started. But you have to start with ut2004-bin-linux-amd64 ! Best regards Hans -- WWW: http://dean.bong.com.au LAN: http://www.bong.com.au EMAIL: [EMAIL PROTECTED] or [EMAIL PROTECTED] ICQ: 16867613 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: unreal tournament 2004
Am Montag, 19. September 2005 14:06 schrieb Dean Hamstead: > indeed i do have the ia32 installed > > ive tried both amd64 and the ia32 binary > > im convinced ive borked something somehow > > Dean > > Jo Shields wrote: > > Dean Hamstead wrote: > >> is there any trick to getting ut2004 working? > >> > >> gltron and tuxracer both work well, ut just hangs when it tries > >> to get into opengl > >> > >> hints? > >> > >> Dean > > > > Do you have ia32 compatability 3d drivers (nvidia-glx-ia32)? > > > > Have you modified the /usr/local/bin/ut2004 script to use amd64 ut2004 > > instead of the default ia32 version? > > > > --Jo > > -- > WWW: http://dean.bong.com.au LAN: http://www.bong.com.au > EMAIL: [EMAIL PROTECTED] or [EMAIL PROTECTED] > ICQ: 16867613 Hello ! Did you install the latest patch ? It is called "ut2004-lnxpatch3339.tar.bz2". Additionally you have to start the ut2004-amd64. Please check your start-script under /usr/local/bin. Maybe you have to edit. By default the 32-bin-version is beeing started. But you have to start with ut2004-bin-linux-amd64 ! Best regards Hans -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: unreal tournament 2004
indeed i do have the ia32 installed ive tried both amd64 and the ia32 binary im convinced ive borked something somehow Dean Jo Shields wrote: Dean Hamstead wrote: is there any trick to getting ut2004 working? gltron and tuxracer both work well, ut just hangs when it tries to get into opengl hints? Dean Do you have ia32 compatability 3d drivers (nvidia-glx-ia32)? Have you modified the /usr/local/bin/ut2004 script to use amd64 ut2004 instead of the default ia32 version? --Jo -- WWW: http://dean.bong.com.au LAN: http://www.bong.com.au EMAIL: [EMAIL PROTECTED] or [EMAIL PROTECTED] ICQ: 16867613 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dupal Opteron on Sarge
Hi Jo, Thanks for the additional info, that was something I didnt know (I should have guessed that). Best regards Nils Valentin Tokyo / Japan http://www.be-known-online.com > [EMAIL PROTECTED] wrote: > >>Forgot to add one thing. >> >>As for the Power Supplies (PS). I tried everything from 400W to 550W and >>came to the conclusion that mostly 400W is fully acceptable for most >>machines (1-4 disks). I know that the shop assistants often recommend >>stronger Power supplies (after all thats their business ;-). >> >>Bear in mind that the heat has to go somewhere ;-) >> >>Best regards >> >>Nils Valentin >>Tokyo / Japan >>http://www.be-known-online.com >> >> >> >>>O.K I guess I have to take the risk that Lord Sauron will flame me for >>>what I say next ;-) >>> >>>I would rather recommend you a 2xCPU MB instead of a single CPU MB >>>You will easily find many 2xCPU MB out there and you may want to >>> consider >>>the following: >>> >>>- most have only 32 bit PCI slots, look for 64 bit PCI slots and the max >>>bus speed (133mhz) >>>- server boards come (optional) with onboard scsi cards >>>- you may still find additionally 4xSATA & IDE on the same MB >>>- (optional) SCSI raid card >>>- (optional) server admin card >>> >>>If you take this all into conclusion you will probably opt for the >>>S2882UG3NR or a similar board from another maker >>>http://www.newegg.com/Product/Product.asp?Item=N82E16813151151 *no scsi >>>http://www.newegg.com/Product/Product.asp?Item=N82E16813151152 * with >>> scsi >>>http://www.newegg.com/Product/Showimage.asp?Mode=&Type=&Image=13-151-152-01.jpg%2C13-151-152-02.jpg%2C13-151-152-03.jpg%2C13-151-152-04.jpg%2C13-151-152-05.jpg&CurImage=13-151-152-02.jpg&Description=TYAN+S2882UG3NR-D+Dual+Socket+940+AMD+8131+Extended+ATX+Server+Motherboard+-+Retail >>> >>>Note the scsi raid card socket (option), white and in the middle left) >>> >>>Having the 64 bit pci bus also allows you to navigate real HEAVY loads >>>between scsi discs and the CPU & Memory. >>> >>>Regarding CPU I would probably choose the 246 or 248 ($240-310) which >>> are >>>slightly cheaper than the 252 and, lets face it, the CPU speed alone >>> wont >>>really change so much on the systems performance. >>> >>>http://www.newegg.com/Product/Product.asp?Item=N82E16819103412 >>>http://www.newegg.com/Product/Product.asp?Item=N82E16819103433 >>> >>>Another concern is the heat those CPU beasts produce (90W each). You may >>>want to find a cooler solution which doesnt grill you or the CPU and >>> also >>>doesnt make you death. (I am sitting next to such a beast right now ;-) >>> >>>Most cooling systems go up to the 246 or 248. Finding one for the 252 >>>might be difficult. I had a hard time finding a cooler which would do >>> his >>>job for the 248 while still allowing me to hear my own voice ;-). >>> >>>I would invest the saved money (compared to the 252) into Memory or SCSI >>>and use the SATA for a backup solution. I prefer SCSI, as for me it has >>>shown to be reliable in the long run and still outperforms most SATA/IDE >>>setups. >>> >>>Note that 2xCPU systems are not twice as fast as single CPU systems. The >>>SMP setup does have a bit of a overhead on your OS, so expect something >>>like 0,8x the speed of a single CPU system. >>> >>>What the 2xCPU system gives you however, is the ability to handle heavy >>>load. It can handle obviously more requests than a single CPU system. It >>>will also take advantage of the hyperthreading bus *communication bus >>>between the CPUs. >>>This means that CPU1 can *borrow Memory from CPU2 if it is required for >>> an >>>application. >>> >>>For the case I would choose the CM Stacker. It does Unfortunately not >>>offer the redundant PS, but you can put 2x PS from the maker of your >>>choice. It wont be hot swabable, you will have to switch the plugs >>>yourself. >>>http://www.newegg.com/Product/Product.asp?Item=N82E1689042 >>> >>>The design is also quite appealing I believe. If you do opt for the CM >>>Stacker make sure to get the Optional Cross floor fan. >>>http://www.newegg.com/Product/Product.asp?Item=N82E16800887011 >>>It will make sure to get the heat out of the case, and that silently. >>> >>>If you prefer to show off the case, there is also a sie window available >>>http://www.newegg.com/Product/Product.asp?Item=N82E16811999760 >>> >>>I hope that I could make some valid points to make your decision easier >>>for your next system. >>> >>>The system above is more expensive than "Lord Saurons" setup but it adds >>> a >>>lot of options and is really a sound system (not to say a beast ;-) >>> >>>I usually buy the parts not all in one go, but as required. >>> >>>Newegg as suggested by "Lord Sauron" does give a good service and I can >>>back him up with that. >>> >>>I currently have 2 Opteron systems online which should be accessable for >>>the public. >>> >>>1) 203.143.127.108 >>>2) 203.143.127.98 (testing) >>>3) 203.143.127.99 (from October) >>> >>>Do you have the possibility to take an optero
Re: Dupal Opteron on Sarge
Hi Joe, Thanks for the additional info, that was something I didnt know (I should have guessed that). Best regards Nils Valentin Tokyo / Japan http://www.be-known-online.com > [EMAIL PROTECTED] wrote: > >>Forgot to add one thing. >> >>As for the Power Supplies (PS). I tried everything from 400W to 550W and >>came to the conclusion that mostly 400W is fully acceptable for most >>machines (1-4 disks). I know that the shop assistants often recommend >>stronger Power supplies (after all thats their business ;-). >> >>Bear in mind that the heat has to go somewhere ;-) >> >>Best regards >> >>Nils Valentin >>Tokyo / Japan >>http://www.be-known-online.com >> >> >> >>>O.K I guess I have to take the risk that Lord Sauron will flame me for >>>what I say next ;-) >>> >>>I would rather recommend you a 2xCPU MB instead of a single CPU MB >>>You will easily find many 2xCPU MB out there and you may want to >>> consider >>>the following: >>> >>>- most have only 32 bit PCI slots, look for 64 bit PCI slots and the max >>>bus speed (133mhz) >>>- server boards come (optional) with onboard scsi cards >>>- you may still find additionally 4xSATA & IDE on the same MB >>>- (optional) SCSI raid card >>>- (optional) server admin card >>> >>>If you take this all into conclusion you will probably opt for the >>>S2882UG3NR or a similar board from another maker >>>http://www.newegg.com/Product/Product.asp?Item=N82E16813151151 *no scsi >>>http://www.newegg.com/Product/Product.asp?Item=N82E16813151152 * with >>> scsi >>>http://www.newegg.com/Product/Showimage.asp?Mode=&Type=&Image=13-151-152-01.jpg%2C13-151-152-02.jpg%2C13-151-152-03.jpg%2C13-151-152-04.jpg%2C13-151-152-05.jpg&CurImage=13-151-152-02.jpg&Description=TYAN+S2882UG3NR-D+Dual+Socket+940+AMD+8131+Extended+ATX+Server+Motherboard+-+Retail >>> >>>Note the scsi raid card socket (option), white and in the middle left) >>> >>>Having the 64 bit pci bus also allows you to navigate real HEAVY loads >>>between scsi discs and the CPU & Memory. >>> >>>Regarding CPU I would probably choose the 246 or 248 ($240-310) which >>> are >>>slightly cheaper than the 252 and, lets face it, the CPU speed alone >>> wont >>>really change so much on the systems performance. >>> >>>http://www.newegg.com/Product/Product.asp?Item=N82E16819103412 >>>http://www.newegg.com/Product/Product.asp?Item=N82E16819103433 >>> >>>Another concern is the heat those CPU beasts produce (90W each). You may >>>want to find a cooler solution which doesnt grill you or the CPU and >>> also >>>doesnt make you death. (I am sitting next to such a beast right now ;-) >>> >>>Most cooling systems go up to the 246 or 248. Finding one for the 252 >>>might be difficult. I had a hard time finding a cooler which would do >>> his >>>job for the 248 while still allowing me to hear my own voice ;-). >>> >>>I would invest the saved money (compared to the 252) into Memory or SCSI >>>and use the SATA for a backup solution. I prefer SCSI, as for me it has >>>shown to be reliable in the long run and still outperforms most SATA/IDE >>>setups. >>> >>>Note that 2xCPU systems are not twice as fast as single CPU systems. The >>>SMP setup does have a bit of a overhead on your OS, so expect something >>>like 0,8x the speed of a single CPU system. >>> >>>What the 2xCPU system gives you however, is the ability to handle heavy >>>load. It can handle obviously more requests than a single CPU system. It >>>will also take advantage of the hyperthreading bus *communication bus >>>between the CPUs. >>>This means that CPU1 can *borrow Memory from CPU2 if it is required for >>> an >>>application. >>> >>>For the case I would choose the CM Stacker. It does Unfortunately not >>>offer the redundant PS, but you can put 2x PS from the maker of your >>>choice. It wont be hot swabable, you will have to switch the plugs >>>yourself. >>>http://www.newegg.com/Product/Product.asp?Item=N82E1689042 >>> >>>The design is also quite appealing I believe. If you do opt for the CM >>>Stacker make sure to get the Optional Cross floor fan. >>>http://www.newegg.com/Product/Product.asp?Item=N82E16800887011 >>>It will make sure to get the heat out of the case, and that silently. >>> >>>If you prefer to show off the case, there is also a sie window available >>>http://www.newegg.com/Product/Product.asp?Item=N82E16811999760 >>> >>>I hope that I could make some valid points to make your decision easier >>>for your next system. >>> >>>The system above is more expensive than "Lord Saurons" setup but it adds >>> a >>>lot of options and is really a sound system (not to say a beast ;-) >>> >>>I usually buy the parts not all in one go, but as required. >>> >>>Newegg as suggested by "Lord Sauron" does give a good service and I can >>>back him up with that. >>> >>>I currently have 2 Opteron systems online which should be accessable for >>>the public. >>> >>>1) 203.143.127.108 >>>2) 203.143.127.98 (testing) >>>3) 203.143.127.99 (from October) >>> >>>Do you have the possibility to take an opter
Re: unreal tournament 2004
Dean Hamstead wrote: is there any trick to getting ut2004 working? gltron and tuxracer both work well, ut just hangs when it tries to get into opengl hints? Dean Do you have ia32 compatability 3d drivers (nvidia-glx-ia32)? Have you modified the /usr/local/bin/ut2004 script to use amd64 ut2004 instead of the default ia32 version? --Jo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
unreal tournament 2004
is there any trick to getting ut2004 working? gltron and tuxracer both work well, ut just hangs when it tries to get into opengl hints? Dean -- WWW: http://dean.bong.com.au LAN: http://www.bong.com.au EMAIL: [EMAIL PROTECTED] or [EMAIL PROTECTED] ICQ: 16867613 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to run qt3*
I have undertood that you are trying to compile a qt3 program and the configure fails, no? try this: ./configure --with-qtdir=/usr/share/qt3 otherewise if the error come from kde headers, probably you need to install some kdelibs-dev, search it: apt-cache search kde dev Regards, Leo A Dilluns 19 Setembre 2005 12:30, Matthias Reinhardt va escriure: > Am Sonntag, 18. September 2005 04:39 schrieb Hamish Moffatt: > > On Sat, Sep 17, 2005 at 08:35:17PM +0100, Matthias Reinhardt wrote: > > > Hi everybody, > > > > > > how can I get qt3* and all the necessary libqt3* workin. > > > I have several problems using appz with qt3 > > > > > > solution: 32bit chroot? > > > > There's no problem with qt3 on amd64. Do you have a particular package > > that is causing problems? > > > > Hamish > > -- > > Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> > > There's always the same failure > checking for KDE... configure: error: > in the prefix, you've chosen, are no KDE headers installed. This will fail. > So, check this please and use another prefix! > > > ls /usr/lib/qt3/ -al > insgesamt 76 > drwxr-xr-x3 root root 4096 2005-09-06 15:27 ./ > drwxr-xr-x 184 root root 65536 2005-09-19 12:10 ../ > drwxr-xr-x5 root root 4096 2005-09-19 11:31 plugins/ > > there are no libs, nor? > > Packages with problems: > libqt3c102-mt > > > -- > greetz Matze. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dupal Opteron on Sarge
[EMAIL PROTECTED] wrote: >Forgot to add one thing. > >As for the Power Supplies (PS). I tried everything from 400W to 550W and >came to the conclusion that mostly 400W is fully acceptable for most >machines (1-4 disks). I know that the shop assistants often recommend >stronger Power supplies (after all thats their business ;-). > >Bear in mind that the heat has to go somewhere ;-) > >Best regards > >Nils Valentin >Tokyo / Japan >http://www.be-known-online.com > > > >>O.K I guess I have to take the risk that Lord Sauron will flame me for >>what I say next ;-) >> >>I would rather recommend you a 2xCPU MB instead of a single CPU MB >>You will easily find many 2xCPU MB out there and you may want to consider >>the following: >> >>- most have only 32 bit PCI slots, look for 64 bit PCI slots and the max >>bus speed (133mhz) >>- server boards come (optional) with onboard scsi cards >>- you may still find additionally 4xSATA & IDE on the same MB >>- (optional) SCSI raid card >>- (optional) server admin card >> >>If you take this all into conclusion you will probably opt for the >>S2882UG3NR or a similar board from another maker >>http://www.newegg.com/Product/Product.asp?Item=N82E16813151151 *no scsi >>http://www.newegg.com/Product/Product.asp?Item=N82E16813151152 * with scsi >>http://www.newegg.com/Product/Showimage.asp?Mode=&Type=&Image=13-151-152-01.jpg%2C13-151-152-02.jpg%2C13-151-152-03.jpg%2C13-151-152-04.jpg%2C13-151-152-05.jpg&CurImage=13-151-152-02.jpg&Description=TYAN+S2882UG3NR-D+Dual+Socket+940+AMD+8131+Extended+ATX+Server+Motherboard+-+Retail >> >>Note the scsi raid card socket (option), white and in the middle left) >> >>Having the 64 bit pci bus also allows you to navigate real HEAVY loads >>between scsi discs and the CPU & Memory. >> >>Regarding CPU I would probably choose the 246 or 248 ($240-310) which are >>slightly cheaper than the 252 and, lets face it, the CPU speed alone wont >>really change so much on the systems performance. >> >>http://www.newegg.com/Product/Product.asp?Item=N82E16819103412 >>http://www.newegg.com/Product/Product.asp?Item=N82E16819103433 >> >>Another concern is the heat those CPU beasts produce (90W each). You may >>want to find a cooler solution which doesnt grill you or the CPU and also >>doesnt make you death. (I am sitting next to such a beast right now ;-) >> >>Most cooling systems go up to the 246 or 248. Finding one for the 252 >>might be difficult. I had a hard time finding a cooler which would do his >>job for the 248 while still allowing me to hear my own voice ;-). >> >>I would invest the saved money (compared to the 252) into Memory or SCSI >>and use the SATA for a backup solution. I prefer SCSI, as for me it has >>shown to be reliable in the long run and still outperforms most SATA/IDE >>setups. >> >>Note that 2xCPU systems are not twice as fast as single CPU systems. The >>SMP setup does have a bit of a overhead on your OS, so expect something >>like 0,8x the speed of a single CPU system. >> >>What the 2xCPU system gives you however, is the ability to handle heavy >>load. It can handle obviously more requests than a single CPU system. It >>will also take advantage of the hyperthreading bus *communication bus >>between the CPUs. >>This means that CPU1 can *borrow Memory from CPU2 if it is required for an >>application. >> >>For the case I would choose the CM Stacker. It does Unfortunately not >>offer the redundant PS, but you can put 2x PS from the maker of your >>choice. It wont be hot swabable, you will have to switch the plugs >>yourself. >>http://www.newegg.com/Product/Product.asp?Item=N82E1689042 >> >>The design is also quite appealing I believe. If you do opt for the CM >>Stacker make sure to get the Optional Cross floor fan. >>http://www.newegg.com/Product/Product.asp?Item=N82E16800887011 >>It will make sure to get the heat out of the case, and that silently. >> >>If you prefer to show off the case, there is also a sie window available >>http://www.newegg.com/Product/Product.asp?Item=N82E16811999760 >> >>I hope that I could make some valid points to make your decision easier >>for your next system. >> >>The system above is more expensive than "Lord Saurons" setup but it adds a >>lot of options and is really a sound system (not to say a beast ;-) >> >>I usually buy the parts not all in one go, but as required. >> >>Newegg as suggested by "Lord Sauron" does give a good service and I can >>back him up with that. >> >>I currently have 2 Opteron systems online which should be accessable for >>the public. >> >>1) 203.143.127.108 >>2) 203.143.127.98 (testing) >>3) 203.143.127.99 (from October) >> >>Do you have the possibility to take an opteron system for a testride ? >>(You might find that some shops might let you try a system) >> >>Best regards >> >>Nils Valentin >>Tokyo / Japan >>http://www.be-known-online.com >> >> >> >> >> > > > > On the topic of power supplies, buy a decent supply from a good brand - far more important than sheer "wattage".
Re: Dupal Opteron on Sarge
Forgot to add one thing. As for the Power Supplies (PS). I tried everything from 400W to 550W and came to the conclusion that mostly 400W is fully acceptable for most machines (1-4 disks). I know that the shop assistants often recommend stronger Power supplies (after all thats their business ;-). Bear in mind that the heat has to go somewhere ;-) Best regards Nils Valentin Tokyo / Japan http://www.be-known-online.com > O.K I guess I have to take the risk that Lord Sauron will flame me for > what I say next ;-) > > I would rather recommend you a 2xCPU MB instead of a single CPU MB > You will easily find many 2xCPU MB out there and you may want to consider > the following: > > - most have only 32 bit PCI slots, look for 64 bit PCI slots and the max > bus speed (133mhz) > - server boards come (optional) with onboard scsi cards > - you may still find additionally 4xSATA & IDE on the same MB > - (optional) SCSI raid card > - (optional) server admin card > > If you take this all into conclusion you will probably opt for the > S2882UG3NR or a similar board from another maker > http://www.newegg.com/Product/Product.asp?Item=N82E16813151151 *no scsi > http://www.newegg.com/Product/Product.asp?Item=N82E16813151152 * with scsi > http://www.newegg.com/Product/Showimage.asp?Mode=&Type=&Image=13-151-152-01.jpg%2C13-151-152-02.jpg%2C13-151-152-03.jpg%2C13-151-152-04.jpg%2C13-151-152-05.jpg&CurImage=13-151-152-02.jpg&Description=TYAN+S2882UG3NR-D+Dual+Socket+940+AMD+8131+Extended+ATX+Server+Motherboard+-+Retail > > Note the scsi raid card socket (option), white and in the middle left) > > Having the 64 bit pci bus also allows you to navigate real HEAVY loads > between scsi discs and the CPU & Memory. > > Regarding CPU I would probably choose the 246 or 248 ($240-310) which are > slightly cheaper than the 252 and, lets face it, the CPU speed alone wont > really change so much on the systems performance. > > http://www.newegg.com/Product/Product.asp?Item=N82E16819103412 > http://www.newegg.com/Product/Product.asp?Item=N82E16819103433 > > Another concern is the heat those CPU beasts produce (90W each). You may > want to find a cooler solution which doesnt grill you or the CPU and also > doesnt make you death. (I am sitting next to such a beast right now ;-) > > Most cooling systems go up to the 246 or 248. Finding one for the 252 > might be difficult. I had a hard time finding a cooler which would do his > job for the 248 while still allowing me to hear my own voice ;-). > > I would invest the saved money (compared to the 252) into Memory or SCSI > and use the SATA for a backup solution. I prefer SCSI, as for me it has > shown to be reliable in the long run and still outperforms most SATA/IDE > setups. > > Note that 2xCPU systems are not twice as fast as single CPU systems. The > SMP setup does have a bit of a overhead on your OS, so expect something > like 0,8x the speed of a single CPU system. > > What the 2xCPU system gives you however, is the ability to handle heavy > load. It can handle obviously more requests than a single CPU system. It > will also take advantage of the hyperthreading bus *communication bus > between the CPUs. > This means that CPU1 can *borrow Memory from CPU2 if it is required for an > application. > > For the case I would choose the CM Stacker. It does Unfortunately not > offer the redundant PS, but you can put 2x PS from the maker of your > choice. It wont be hot swabable, you will have to switch the plugs > yourself. > http://www.newegg.com/Product/Product.asp?Item=N82E1689042 > > The design is also quite appealing I believe. If you do opt for the CM > Stacker make sure to get the Optional Cross floor fan. > http://www.newegg.com/Product/Product.asp?Item=N82E16800887011 > It will make sure to get the heat out of the case, and that silently. > > If you prefer to show off the case, there is also a sie window available > http://www.newegg.com/Product/Product.asp?Item=N82E16811999760 > > I hope that I could make some valid points to make your decision easier > for your next system. > > The system above is more expensive than "Lord Saurons" setup but it adds a > lot of options and is really a sound system (not to say a beast ;-) > > I usually buy the parts not all in one go, but as required. > > Newegg as suggested by "Lord Sauron" does give a good service and I can > back him up with that. > > I currently have 2 Opteron systems online which should be accessable for > the public. > > 1) 203.143.127.108 > 2) 203.143.127.98 (testing) > 3) 203.143.127.99 (from October) > > Do you have the possibility to take an opteron system for a testride ? > (You might find that some shops might let you try a system) > > Best regards > > Nils Valentin > Tokyo / Japan > http://www.be-known-online.com > > > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to run qt3*
Matthias Reinhardt wrote: Am Sonntag, 18. September 2005 04:39 schrieb Hamish Moffatt: On Sat, Sep 17, 2005 at 08:35:17PM +0100, Matthias Reinhardt wrote: Hi everybody, how can I get qt3* and all the necessary libqt3* workin. I have several problems using appz with qt3 solution: 32bit chroot? There's no problem with qt3 on amd64. Do you have a particular package that is causing problems? Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> There's always the same failure checking for KDE... configure: error: in the prefix, you've chosen, are no KDE headers installed. This will fail. So, check this please and use another prefix! ls /usr/lib/qt3/ -al insgesamt 76 drwxr-xr-x3 root root 4096 2005-09-06 15:27 ./ drwxr-xr-x 184 root root 65536 2005-09-19 12:10 ../ drwxr-xr-x5 root root 4096 2005-09-19 11:31 plugins/ there are no libs, nor? Packages with problems: libqt3c102-mt So by "run qt3*" you mean "install development headers for QT" Try libqt3-headers, or the somewhat more useful libqt3-mt-dev -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to run qt3*
Am Sonntag, 18. September 2005 04:39 schrieb Hamish Moffatt: > On Sat, Sep 17, 2005 at 08:35:17PM +0100, Matthias Reinhardt wrote: > > Hi everybody, > > > > how can I get qt3* and all the necessary libqt3* workin. > > I have several problems using appz with qt3 > > > > solution: 32bit chroot? > > There's no problem with qt3 on amd64. Do you have a particular package > that is causing problems? > > Hamish > -- > Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> There's always the same failure checking for KDE... configure: error: in the prefix, you've chosen, are no KDE headers installed. This will fail. So, check this please and use another prefix! ls /usr/lib/qt3/ -al insgesamt 76 drwxr-xr-x3 root root 4096 2005-09-06 15:27 ./ drwxr-xr-x 184 root root 65536 2005-09-19 12:10 ../ drwxr-xr-x5 root root 4096 2005-09-19 11:31 plugins/ there are no libs, nor? Packages with problems: libqt3c102-mt -- greetz Matze. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dupal Opteron on Sarge
O.K I guess I have to take the risk that Lord Sauron will flame me for what I say next ;-) I would rather recommend you a 2xCPU MB instead of a single CPU MB You will easily find many 2xCPU MB out there and you may want to consider the following: - most have only 32 bit PCI slots, look for 64 bit PCI slots and the max bus speed (133mhz) - server boards come (optional) with onboard scsi cards - you may still find additionally 4xSATA & IDE on the same MB - (optional) SCSI raid card - (optional) server admin card If you take this all into conclusion you will probably opt for the S2882UG3NR or a similar board from another maker http://www.newegg.com/Product/Product.asp?Item=N82E16813151151 *no scsi http://www.newegg.com/Product/Product.asp?Item=N82E16813151152 * with scsi http://www.newegg.com/Product/Showimage.asp?Mode=&Type=&Image=13-151-152-01.jpg%2C13-151-152-02.jpg%2C13-151-152-03.jpg%2C13-151-152-04.jpg%2C13-151-152-05.jpg&CurImage=13-151-152-02.jpg&Description=TYAN+S2882UG3NR-D+Dual+Socket+940+AMD+8131+Extended+ATX+Server+Motherboard+-+Retail Note the scsi raid card socket (option), white and in the middle left) Having the 64 bit pci bus also allows you to navigate real HEAVY loads between scsi discs and the CPU & Memory. Regarding CPU I would probably choose the 246 or 248 ($240-310) which are slightly cheaper than the 252 and, lets face it, the CPU speed alone wont really change so much on the systems performance. http://www.newegg.com/Product/Product.asp?Item=N82E16819103412 http://www.newegg.com/Product/Product.asp?Item=N82E16819103433 Another concern is the heat those CPU beasts produce (90W each). You may want to find a cooler solution which doesnt grill you or the CPU and also doesnt make you death. (I am sitting next to such a beast right now ;-) Most cooling systems go up to the 246 or 248. Finding one for the 252 might be difficult. I had a hard time finding a cooler which would do his job for the 248 while still allowing me to hear my own voice ;-). I would invest the saved money (compared to the 252) into Memory or SCSI and use the SATA for a backup solution. I prefer SCSI, as for me it has shown to be reliable in the long run and still outperforms most SATA/IDE setups. Note that 2xCPU systems are not twice as fast as single CPU systems. The SMP setup does have a bit of a overhead on your OS, so expect something like 0,8x the speed of a single CPU system. What the 2xCPU system gives you however, is the ability to handle heavy load. It can handle obviously more requests than a single CPU system. It will also take advantage of the hyperthreading bus *communication bus between the CPUs. This means that CPU1 can *borrow Memory from CPU2 if it is required for an application. For the case I would choose the CM Stacker. It does Unfortunately not offer the redundant PS, but you can put 2x PS from the maker of your choice. It wont be hot swabable, you will have to switch the plugs yourself. http://www.newegg.com/Product/Product.asp?Item=N82E1689042 The design is also quite appealing I believe. If you do opt for the CM Stacker make sure to get the Optional Cross floor fan. http://www.newegg.com/Product/Product.asp?Item=N82E16800887011 It will make sure to get the heat out of the case, and that silently. If you prefer to show off the case, there is also a sie window available http://www.newegg.com/Product/Product.asp?Item=N82E16811999760 I hope that I could make some valid points to make your decision easier for your next system. The system above is more expensive than "Lord Saurons" setup but it adds a lot of options and is really a sound system (not to say a beast ;-) I usually buy the parts not all in one go, but as required. Newegg as suggested by "Lord Sauron" does give a good service and I can back him up with that. I currently have 2 Opteron systems online which should be accessable for the public. 1) 203.143.127.108 2) 203.143.127.98 (testing) 3) 203.143.127.99 (from October) Do you have the possibility to take an opteron system for a testride ? (You might find that some shops might let you try a system) Best regards Nils Valentin Tokyo / Japan http://www.be-known-online.com -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: http://http.us.debian.org/debian-amd64/debian/ gone?
On (18/09/05 13:23), Greg Madden wrote: > On Friday 16 September 2005 01:22 am, Clive Menzies wrote: > > On (15/09/05 15:56), lordSauron wrote: > > > Somewhat related, but not really, (I'm shy about starting new > > > threads), I was wondering what mirrors there are for amd64-k8 or > > > otherwise amd64 compatible packages. Right now I'm using the > > > following mirrors: > > > > > > deb http://linux.csua.berkeley.edu/debian/ stable main > > > deb-src http://linux.csua.berkeley.edu/debian/ stable main > > > > > > deb http://security.debian.org/ stable/updates main > > > > deb http://ftp.de.debian.org/debian-amd64/debian/ sid main contrib > > non-free > > deb-src http://ftp.de.debian.org/debian-amd64/debian/ sid main > > contrib non-free > > > > #deb http://ftp.nl.debian.org/debian-amd64/debian/ sid main contrib > > non-free > > #deb-src http://ftp.nl.debian.org/debian-amd64/debian/ sid main > > contrib non-free > > > > omit .de (.nl) The above is for sid/unstable substitute etch/testing > > or sarge/stable > > > > > and if you could tell me where to go to get specifically amd64 > > > packages I'd be much appreciative! > > > > > > Also, if you know how I could specify to select from a variety of > > > packages (like look at both stable and testing, for instance) I'd > > > love to know that as well. I very new to Linux, so please just > > > bear with me. > > > > If you're new to Debian, I would stick with one release FTTB; once > > you're comfortable with it, look at apt-pinning. > > This thread has me confused. AFAIKT, the only place to get debian-amd64 > packages is from: http://amd64.debian.net/ or a mirror : > http://amd64.debian.net/README.mirrors.txt. Well those above are from my sources.list and I'm running sid amd64 :) Regards Clive -- www.clivemenzies.co.uk ... ...strategies for business -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dupal Opteron on Sarge
Ron Johnson wrote: On Sun, 2005-09-18 at 13:18 +1000, Hamish Moffatt wrote: On Sat, Sep 17, 2005 at 06:49:55PM -0700, lordSauron wrote: pentium 4s use a 21 stage pipeline or something like that... so they take approximately 21 clock cycles to get anything done. AMD uses about 7 stages (or something in that neighbourhood) so if you divide 2.8 by 21 and 2.0 (my Athlon64) by 7, you get a really interesting breakdown. You'll certainly find a HUGE increase in performance, That's a terrible simplification. Yes, it takes longer to get the first Not only is it a simplification, it's wrong. result (21 cycles versus 7) but the idea of the pipeline is that you can get a result every clock cycle after that. But when you context-switch or branch, the pipeline gets dirty, and the new process needs to fill up the pipeline. Short pipelines like in Athlon & G4 are easier on branching, but other techniques like speculative fetching and OOE mitigate that somewhat. And then, deep pipelines let you ramp up the clock much easier than do short pipelines. Don't know why, though. You can ramp up the clock speed on a deep pipeline, because each stage in the pipeline do very little. Therefore, it can be done faster than the more complex stages found in a shorter pipeline doing the same job. Unfortunately, ability to ramp up the clock doesn't help when a faster clock becomes necessary just to keep up. We all know how an opteron beats any pentium with the _same_ clock rating by a wide margin. A 21-stage pipeline _is_ too deep - on average, x86 instructions have a branch for every 7th instruction, and each branch may invalidate that deep pipeline. A shorter pipeline have less of these problems. The latency is higher but the throughput is also higher (more clock cycles per second). Yes, except when troughput is ruined by latency from all those branches. People try fixing this by unrolling loops, but then they use up the instruction cache instead. A good pipeline should be short enough to be (almost) full most of the time. Otherwise it is going to loose, no matter what clock speed. Helge Hafting -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
