Older blog entries for crudman (starting at number 14)

The Trouble with (SMP) Tribbles - AKA Rant

{sigh} Where to begin?

Recently I've been trying to make my main home machine (consisting of an Asus P2B-D, rev D06, and two Pentium 3 733mhz's), work for more than 5 hours without a hard-lock. No dice so far.

After exhausting all other options (cooling, new fans, slockets, power supply, etc), I've finally decided that the problem is with the board itself. Each CPU runs at 133 Front Side Bus (a purchasing decision I now regret, due to the CPUs having their multipliers locked), which the board can support. Unfortunately, it's only stable at 66/100 FSB (which forces the clock speed down).


It's not the first time I've built a SMP system either. I was also foolish enough to purchase the bastard of all SMP motherboards (heh, excluding the Tyan Tiger 100 apparently): The Abit BP6.

The motherboard manual explicitly stated the dual Socket 370 ability was "experimental" (a pseudo-legal disclaimer for Abit's technical support). Run it at your own risk. Being one of the only SMP boards available at the time, most people were happy to live with that.

Getting this board stable took three months of my life. What started out as a simple upgrade from a P2/300, ended up in the occasional fit of rage. I've outlined the bulk of my experiences here.

In disgust, I canned the board (decided to keep it instead of a RMA for some reason) and purchased my current board, the Asus P2B-D. Finally, stability. Ran like a dream (at Celeron FSB: 66mhz). Amazed by the kernel compilation speed and MP3 ripping, etc, etc. Decided never to buy another Abit product again.

A few months later, I read about the "EC10" modification that's reported on the bp6.com forums. Basically, someone discovered that a recent revision of the BP6 board came with a higher rated capacitor (in the EC10 position). An additional capacitor was wired in parallel to the existing one, and it fixed the bulk of the BP6's stability issues (especially voltage discrepancies).

Performed the EC10 fix on my BP6. Bang. I was able to do a complete e2fsck. Repeatedly. And other things. Standard things. It's now my file/print/other server to this day. Renewed my decision to never buy another Abit product.

As time went on, more demanding applications (cough, games, cough) required more horsepower, so the current P3's were purchased. To accommodate them, the slockets had to be replaced, and PC-133 memory was required (due to the 'locked multipliers' issue).

With both CPUs at 133 FSB, my box hard-locks (no response from mouse, keyboard or otherwise, including 2 second sound repetition) anywhere from 15 minutes to 5 hours after initial bootup. Which brings us back to the present.

What is it about SMP boards? Why are they more prone to experience problems than their solo-CPU counterparts? It could be argued that all motherboards require the odd BIOS update to fix ongoing issues (eg. controller problems, compatibility, etc), but SMP boards are notoriously bad for having all kinds of configuration issues (eg. sufficient power supply, proper cooling, hardware/software that's SMP capable) together with all the other hassles that are associated with solo-CPU boards.


After a bit more tweaking, its uptime is now 4 hours and 55 minutes. Should it fall over again, I'll think about acquiring a more capable SMP board. My choices are the Abit VP6 (as yet unreleased and untested, wow, BP6 all over again), a Tyan Tiger 133 or MSI 694D.

The VP6? Fool me once, shame on me. Fool me twice...

The pursuit of goodness

The reiserfs partitions continue to work. Grande. Data has been moved back, all is well. I'm a bit reluctant to try the 3.6.x branch at this stage (it's stamped 'development' for a reason, heh).

Recently received slocket-mounted 'Tornado' fans to finally suck the heat out of my main SMP setup (two PIII/733's). Still get the odd hard-lock, but after 2hrs 11mins uptime, things are looking up. I'm happy with the cooling setup, just need to tweak a few more things.

Getting up early to obtain some more parts (one of the rare times I'm able to), then back to business (esp. writing). Annual review coming up early next week, this should be fun. :)

jbowman: I sent you some info about stabilizing your BP6, did you receive this?

Mr. Sandman

There's something about having a four hour sleep debt that numbs various parts of your consciousness. Microsleeps aren't fun.

Spent the past few days converting the filesystems on a server and workstation to reiserfs. Dynamic inode allocation is grand. The reiserfs-debian rescue/root disks help immensely for new Debian installations as well.

For some reason, I set up some recently decommissioned machines at work (P166s) to function as distributed.net clients. I helped a friend install the win32 version of the client, he was presented with the standard configuration dialog. I quote:

"Argh! Text-based configuration! This is Windows 2000 damnit!"

The most unfortunate part is that he's a tech. I guided him through the relevant options to allow communication through a proxy server, but I just write this off as the "mindset" (ie. total GUI interaction) that I see all too often. Wonder what would happen if I hid the mouse...

Ah, it's time to funk

Wow, Sierra/Dynamix announced they've contracted out the Linux client development of Tribes 2 to Loki. We all knew that a Linux server port would be available, but goddamn! Kudos to all involved.

Decided to put these SMP systems to good use, by running distributed.net clients. It's good to crack some blocks. It'll be a good test of my now seemingly-stable BP6 system (Dual Celeron 500mhz, not o/c'ed, with EC10 modification performed). 17 hours so far...

Hmm... Quake III can be executed with a '-nosplash' switch (automatic skip past the id logo movie). Good to know.

Best get back to the 15 odd things that require attention. Some minor, some major. Work (when it resumes next week) is only going to complicate things further. Arg.

Pop Dog: The dog that refreshes

Ah, I've been certified as a Journeyer. My thanks to those involved, guess I'd best make it worth everyone's while.

Recently returned from a trip to Phillip Island (extreme south of Victoria). A lot busier down there than I thought, Cowes is a nice place to visit. Had a coffee at "MadCowes" of all places.

The emulation article is just about done. The Playstation Emulator (FPSE, mentioned in an earlier entry) can't be tested, unless I can get a hold of some titles in the next few days. Considering there are only a few games that be properly played with it, it's becoming more of a non-issue.

The main file server (Netware) at work recently experienced a few issues. Seems a user decided to go nuts in creating/copying many different subdirectories, to the extent that Netware's own tools (FILER) wouldn't correct the problem. Connecting from a DOS box and running a DELTREE command would cause a NLM on the server to stop functioning, and the CPU utilization to sky-rocket (all without any actual progress). Fun stuff.

The solution? Fire up my Debian work box, insmod the IPX and NCP modules, ncpmount the troublesome volume, cd to the specific user account in question, and 'rm -rf' the subdirectory that wouldn't play fair. Two seconds later, it's gone. And my co-workers wonder why I use a UNIX varient in a win32 network. :)

A friend of mine is interested in building a small file/print/web server for home use, he's secured a P100 from a variety of parts lying about at his place of employ. Getting a few copies of RH7 made to help expediate the process.

icemonk: My correspondence with the International Olympic Committee (IOC) isn't paying off. They say LAN games of Counter-Strike won't even be considered for inclusion as an official Olympic event, until the year 2032, or when 'hell freezes over' (whichever comes first). Damn.

"Implosion"? I thought you said...

Spent a bit of time today trying to help icemonk get his game box working with the Nvidia 0.9-5 drivers. Aside from a few strange I840 messages during compilation (it's a BX motherboard), everything seemed rosy. At least until UT was installed and started. Hard locks galore.

Anyone have a working Linux UT (428a) setup and using RH7 beta? Just wondering if the very recent glibc that's included is to blame (noticed that Loki put out a patch for SC3K to correct a similiar issue).

There's quite a few arcade/console emulators about for Linux, up to 17 and counting. The guide I'm working on should be grande once it's finished (real soon now). That last pistol-whipping hurt. :(

Here's something: FPSE (The Free PlayStation Emulator) has Linux binaries and source now available. I'm completely lacking in any PlayStation titles, if anyone is able to test this out and report back any findings (especially screenshots), it'd be much appreciated.

crudman at penguinpowered dot com

Still trying to finish 101 various things. And the toaster decides to die on me. Damn.

Been thinking about placing the good ol' Voodoo 1 card back into my system, just to test out the glide functions in gsnes9x and a few other things (UT, Descent III, etc). Should be decent with a reasonably low resolution.

Mozilla M18 (okay, the nightly build of it anyway) is impressive to say the least. Uraeus posted up a write-up of the major changes over at LP.

Downloading the Debian woody packages of Helix Code as I write. Considering I've been using Blackbox for quite a damn long while now, this'll come as quite a change. The last time I tried HC was with a SuSE 6.4 system, and GDM was slightly screwed to say the least. Hope history doesn't repeat.

15 Sep 2000 (updated 15 Sep 2000 at 10:55 UTC) »

Choose life. Choose crud. Choose to turn off the Olympics. Choose to ignore the television station that's milking the sucker for all it's worth. Choose to ignore the one commentator who just happened to be part of the torch relay when it passed by his home town. Choose to locate, configure, test and write about emulators available for the Linux platform. Choose between nestra, tuxnes and DarcNES and attempt to work out which of them actually work correctly. Choose Ogg Vorbis. Choose to post a diary entry. Choose to avoid a pistol-whipping from katzj.

Choose your future.
Choose to advogato.

The mouse arrived today. Oh yeah. Smooth, but a little tricky to initially adjust to.

Shogo! The Linux port has been in development by Hyperion for quite some time, sounds like they're making good progress (OpenGL renderer, sound, etc). I can remember playing this resource-hogging game on a Voodoo 1 (and a short-lived i740) quite a while ago. Must set up a Shogo server when this gets released.

Discovered that SoF can indeed take advantage of S3TC (Display Options -> Texturing). Not nearly as a dramatic visual improvement as UT, but it does provide a noticable performance improvement. Less churn = more speed. Or something.

Slashdot has an "Internet Explorer" (with icon) section for their news now? Why? Surely a news item like this would've fallen under the categories of either security or privacy.

claudio: Ah, you'll be pleased to know that Gimp takes screenshots of GL windows correctly over here. That should provide you with some hope. :)


Thanks to a rather, well, stupid use of an extended IDE cable a while back (moral of the story, don't use extended IDE cabling), my local collection of MP3 files is now slightly corrupted. Thought this would now be a good time to start using the Ogg Vorbis audio codec. It's just crunchier.

Incredibily easy to setup, used the apt-get feature of Debian to retrieve the "vorbis-tools", "ogg123" and "xmms-vorbis" packages. Encoders and decoders all neatly installed. Found quite the useful page for configuring oggenc to work with Grip also. Need to slightly adjust it though.

The video card can indeed be overclocked, but the utility used to reprogram a BIOS file (BIOSMOD) requires a script file specific to the card's chipset (eg. TNT2, GF1/GF2). No script for GF2-MX at this stage, damn.

claudio: With regards to black pixels appearing when building mpeg movies, is this while using the new 0.9-5 drivers? I also had a severe black pixel problem with xmame.xgl, the new ones indicate that "color palette problems" are now fixed (from changelog).

Time to finish the emulation article. Right after sleep.

5 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!