timdoug's interesting tidbits

Little bits of technical documentation and such. Hopefully helpful.


Installing Debian in xhyve

xhyve is a nice and lightweight hypervisor for macOS. Getting Debian running on it requires a bit of effort though; the following instructions result in a functional installation with working networking.

  1. git clone https://github.com/mist64/xhyve.git
  2. Apply this patch so there's enough space for the installer ramdisk (thanks to this PR for the tip):
  3. diff --git a/src/firmware/kexec.c b/src/firmware/kexec.c
    index 61aeebb..2a5ce67 100644
    --- a/src/firmware/kexec.c
    +++ b/src/firmware/kexec.c
    @@ -184,7 +184,7 @@ kexec_load_ramdisk(char *path) {
            sz = (size_t) ftell(f);
            fseek(f, 0, SEEK_SET);
    -       ramdisk_start = ALIGNUP((kernel.base + kernel.size), 0x1000ull);
    +       ramdisk_start = ALIGNUP((kernel.base + kernel.size + 32*1024*1024), 0x1000ull);
            if ((ramdisk_start + sz) > memory.size) {
                    /* not enough memory */
  4. make
  5. Grab the Debian installer kernel and initrd. I use the daily snapshots, but substitute accordingly.
    curl https://d-i.debian.org/daily-images/amd64/daily/netboot/debian-installer/amd64/initrd.gz -o initrd.gz-installer
    curl https://d-i.debian.org/daily-images/amd64/daily/netboot/debian-installer/amd64/linux -o linux-installer
  6. Create an empty disk image: dd if=/dev/zero of=debian.img bs=1m count=8192
  7. Grab a UUID so the VM comes up with the same IP address every launch: python -c 'import uuid; print uuid.uuid4()'
  8. Run the Debian installer, changing options accordingly (1G is RAM size, the arguments to virtio-blk and the kexec firmware are paths): sudo ./build/xhyve -A -H -U <<<YOUR UUID HERE>>> -m 1G -s 0,hostbridge -s 1,lpc -s 2,virtio-blk,debian.img -s 3,virtio-net -l com1,stdio -f "kexec,linux-installer,initrd.gz-installer,earlyprintk=serial console=ttyS0" Feel free to skip bootloader installation, since we use xhyve's kexec implementation.
  9. We now need to grab the non-installer kernel and initrd out of the disk image.
    1. Boot into the installer's recovery mode by passing rescue/enable=true: sudo ./build/xhyve -A -H -U <<<YOUR UUID HERE>>> -m 1G -s 0,hostbridge -s 1,lpc -s 2,virtio-blk,debian.img -s 3,virtio-net -l com1,stdio -f "kexec,linux-installer,initrd.gz-installer,earlyprintk=serial console=ttyS0 rescue/enable=true"
    2. Use /dev/vda1 as the root filesystem and execute a shell there (both the defaults), then type bash to get a sane one.
    3. apt-get install netcat
    4. On the host run nc -l 8000 >vmlinuz, and on the VM run nc 8000 </boot/vmlinuz*. You'll have to manually ctrl-C after the (essentially instant) transfer is complete.
    5. Do the same for the initrd.img in /boot.
    6. Verify checksums and sanity otherwise:
      host$ file vmlinuz initrd.img
      vmlinuz:    Linux kernel x86 boot executable bzImage, version 4.9.0-3-amd64 (debian-kernel@lists.debian.org) #1 SMP Debian 4., RO-rootFS, swap_dev 0x3, Normal VGA
      initrd.img: gzip compressed data, last modified: Thu Jun  1 16:23:39 2017, from Unix
    7. Exit out of bash and the chroot, then "reboot" the system.
  10. Start your new VM, being sure to change the kernel/initrd paths and set the root partition's location accordingly: sudo ./build/xhyve -A -H -U <<<YOUR UUID HERE>>> -m 1G -s 0,hostbridge -s 1,lpc -s 2,virtio-blk,debian.img -s 3,virtio-net -l com1,stdio -f "kexec,vmlinuz,initrd.img,earlyprintk=serial console=ttyS0 root=/dev/vda1"
  11. Success! Install an SSH server, get work done, etc.
Other notes:

[/osx] permanent link


Using the TCP MD5 Signature Option in Linux

It's entirely undocumented, for good reason. In case you also have a peculiar interest in RFC2385, this is what you need to do:

#include <linux/tcp.h>


struct tcp_md5sig md5sig;
char *key = "foobar";

/* addr is the struck sockaddr * passed to connect(2), from getaddrinfo() or otherwise */
memcpy(&md5sig.tcpm_addr, addr, sizeof addr);

md5sig.tcpm_keylen = strlen(key);
memcpy(&md5sig.tcpm_key, key, md5sig.tcpm_keylen);

setsockopt(sockfd, IPPROTO_TCP, TCP_MD5SIG, &md5sig, sizeof md5sig);
Not actually that hard. A successful application, with -M and the key passed to tcpdump, will show md5valid in the TCP options. An incorrect key will show md5invalid.

[/coding] permanent link


Using TLS v1.2 with OpenVPN 2.3.4

It's disabled by default, and if you add e.g. tls-cipher TLS-DHE-RSA-WITH-AES-256-GCM-SHA384 (a TLS v1.2 cipher) to your server config it silently breaks. The solution is to add tls-version-min 1.2 to both the server and client config, and then it works like a charm. OpenVPN 2.3.3 added support for v1.2, so this won't work with anything older than 2.3.3.

[/general] permanent link


How to Watch Sopcast/Ustream Feeds with VLC on Mac OS X

For certain international sports, the two most popular kinds of streams I've come across are Sopcast and Ustream. Both require a bit of work to set up, but they tend to provide 720p-quality and in VLC they don't contain the endless barrage of advertisements or cause Flash Player-related CPU hogging. I suggest you set everything up well in advance and test with non-sport streams.


  1. Install Xcode through the App Store. This is needed to build the following tools.
  2. Download VLC and make sure to copy it to /Applications.
  3. Install Homebrew. The instructions on the page should be self-explanatory.
  4. Open Terminal.app and type:
    1. brew install rtmpdump
    2. sudo bash -c "CFLAGS=-Qunused-arguments CPPFLAGS=-Qunused-arguments easy_install cffi"
    3. sudo bash -c "CFLAGS=-Qunused-arguments CPPFLAGS=-Qunused-arguments easy_install python-librtmp"
    4. sudo easy_install livestreamer
    These installation instructions you only need to do once.
  5. Once you've got a Ustream URL, open Terminal.app and type livestreamer http://www.ustream.tv/eukanuba substituting the URL appropriately. It'll spit out a list of quality settings.
  6. With that quality setting, now type livestreamer http://www.ustream.tv/eukanuba 480p (note the suffixed quality value) and it'll open VLC and start your stream. Make sure to leave Terminal.app open, or your stream will stop.


  1. Download the Mac .dmg from here.
  2. Mount the downloaded disk image but don't do anything with the application.
  3. Open Terminal.app and type mkdir -p ~/bin && cp /Volumes/SopCast/SopCast.app/Contents/Resources/binaries/m32/sp-sc-auth ~/bin. You can now unmount and remove the disk image. Everything up to now you only need to do once.
  4. When you've found a stream, open Terminal.app again and type ~/bin/spsc-auth sop://broker.sopcast.com:3912/xxxxxx 3000 3001 (substituting the Sopcast URL appropriately) and leave it running. If you quit Terminal.app it'll stop the stream.
  5. Open VLC and go to File -> Open Network.... In the URL field type http://localhost:3001 and click Open; you should be good to go.

[/osx] permanent link


Setting up IPv6 with Sonic.net and an OpenWRT Router

This uses the newer 6rd mechanism instead of the 6in4 tunnels, so other documentation you enounter may inapplicable. I'm using nightly builds; YMMV with stable builds.

  1. Enable "LAN Subport" on the equipment provided by Sonic.net for your OpenWRT router (might work with double NAT, haven't tried)
  2. On your router: opkg update && opkg install 6rd (if it complains about version conflicts, try flashing the most recent build)
  3. Make the wan6 entry in your /etc/config/network look like so:
    config interface 'wan6'
            option proto '6rd'
            option peeraddr ''
            option ip6prefix '2602:240::'
            option ip6prefixlen '28'
  4. /etc/init.d/networking restart
  5. Disconnect and reconnect your machine, and IPv6 autoconfiguration should be good to go.
Addresses were taken from this post on the Sonic.net forums, and configuration for OpenWRT from here.

[/general] permanent link


The smallest preseed file for an entirely automated Ubuntu 12.04 Precise Pangolin install

Pass these options on the kernel command line: auto=true priority=critical url=http://your.server/foo.seed, and use this preseed file:

d-i debian-installer/locale string en_US.UTF-8

d-i passwd/root-login boolean true
d-i passwd/make-user boolean false
d-i passwd/root-password password changemeplease
d-i passwd/root-password-again password changemeplease

d-i partman-auto/method string regular
d-i partman-auto/choose_recipe select atomic
d-i partman/choose_partition select finish
d-i partman/confirm boolean true
d-i partman/confirm_nooverwrite boolean true

d-i finish-install/reboot_in_progress note
This will assuredly do thinks you don't want / care about, to make sure to amend it accordingly.

[/ubuntu] permanent link


I can't install gcc on OpenLogic CentOS!

Yeah, someone really dropped the ball here. If you see this when you try to yum install gcc:

--> Finished Dependency Resolution
Error: Package: glibc-headers-2.12-1.80.el6_3.7.x86_64 (updates)
           Requires: kernel-headers
Error: Package: glibc-headers-2.12-1.80.el6_3.7.x86_64 (updates)
           Requires: kernel-headers >= 2.2.1
 You could try using --skip-broken to work around the problem
 You could try running: rpm -Va --nofiles --nodigest
Then try doing this: sudo yum --disableexcludes=main install kernel-headers-*.el6.openlogic.x86_64

I don't know who's the idiot who put exclude=kernel* in /etc/yum.conf to break installing gcc, but...you're an idiot.

[/gcc] permanent link


How to install and use Errbit through nginx & Passenger

Capistrano scares the shit out of me. I'd rather know what I'm deploying, rather than rely on some foreign script, thank you very much. To install Errbit and deploy with Passenger through nginx, try this:

git clone https://github.com/errbit/errbit.git
rake errbit:copy_configs
vi config/config.yml # and change hostnames / SMTP settings
vi config/mongoid.yml # and change production mongo settings
echo "Errbit::Application.config.secret_token = '$(bundle exec rake secret)'" > config/initializers/secret_token.rb
RAILS_ENV=production rake assets:precompile
RAILS_ENV=production rake db:mongoid:create_indexes
RAILS_ENV=production rake db:seed
Then add something like the following to your nginx config:
server {
    listen 8080;
    server_name errbit.example.com;
    root /wherever/errbit/public;
    passenger_enabled on;
    rails_env production;

[/general] permanent link


Use tcpdump to download arbitrary Flash videos

I came across an interesting video on the web that I wanted to watch offline, but it played through a Flash application, there were no download links, grabbing the stream by looking through the HTML/DOM was nontrivial, and the youtube-dl mainstay didn't work. tcpdump to the rescue!

  1. tcpdump -v -i <interface> -w output.cap
  2. Load the video up and start playing. Sadly, this has to be done in real-time.
  3. When done, tcpflow -r output.cap
  4. The largest file should be your video; use an editor to strip the HTTP headers from the beginning.
  5. If all goes well, you should have a video.flv. For extra credit, use ffmpeg -i video.flv to see if it's H.264/AAC. If so, use ffmpeg -i video.flv -acodec copy -vcodec copy video.mp4 for lossless container conversion to a standard MPEG-4 Part 14 file.
Thanks to this post for the tip.

[/general] permanent link


How to send email through Gmail / Google Apps Email with the Amazon Linux AMI on EC2

First, install the Heirloom mailx client with yum install mailx. The NSS certificates included with the AMI are either out of date or incomplete, so you've got to grab the Equifax root cert and set it up yourself, like this:

wget https://www.geotrust.com/resources/root_certificates/certificates/Equifax_Secure_Certificate_Authority.cer
certutil -d ~/.mailcerts/ -A -t TC -n "Equifax Secure Certificate Authority" -i Equifax_Secure_Certificate_Authority.cer
Then set up your ~/.mailrc like this:
set smtp-use-starttls
set smtp=smtp://smtp.gmail.com:587
set smtp-auth=login
set smtp-auth-user="your_username@example.com"
set smtp-auth-password="your_password"
set from="your_username@example.com"
set nss-config-dir="~/.mailcerts"
set ssl-verify=warn
That last "ssl-verify=warn" line looks pretty dangerous, and it definitely could be. I needed it because otherwise mailx would bark with Comparing DNS name: "smtp.gmail.com" Continue (y/n)? and...giving up, instead of accepting any input. For whatever reason my Debian box works fine without this line. Also, one could use this line without setting up the root certificate appropriately, but that's playing a bit too fast and loose for me.

With this configuration you should be good to go with: echo "body here" | mailx -s "Subject here" recipient@example.com

[/aws] permanent link


My 2007-era MacBookPro3,1 doesn't fully go to sleep / cannot wake from sleep with 10.8!

Yep, I had a reason to break out my old MBP and install the newest version of OS X on it. It works wonderfully for a five year old box (the SSD sure helps!), except for not being able to sleep properly. What worked for me was disabling "safe sleep" a la this hint.

sudo pmset -a hibernatemode 0
sudo nvram "use-nvramrc?"=false
Then restart. You can also remove /private/var/vm/sleepimage to save some space, which is nice.

[/osx] permanent link


How to install a fully patched and up-to-date lighttpd on Debian unstable that passes Qualys SSL Labs' tests

...well, except for the cipher suites that require an OpenSSL >= 1.0.0, but that's an adventure for another day.

wget http://redmine.lighttpd.net/attachments/download/1395/ssl-compression.diff
wget http://ftp.us.debian.org/debian/pool/main/l/lighttpd/lighttpd_1.4.31-3.debian.tar.gz
wget http://ftp.us.debian.org/debian/pool/main/l/lighttpd/lighttpd_1.4.31.orig.tar.gz
tar xvzf lighttpd_1.4.31.orig.tar.gz
cd lighttpd-1.4.31/
tar xvzf ../lighttpd_1.4.31-3.debian.tar.gz
vi debian/control [and get rid of the dpkg-dev version dependency]
vi debian/rules [and get rid of the "export=config" line and the previous backslash]
patch -p1 <../ssl-compression.diff
debuild -us -uc
[then install all of the dependencies it barks about and try again...]
cd ..
sudo dpkg -i lighttpd_1.4.31-3_amd64.deb
sudo /etc/init.d/lighttpd restart
And make sure these options are in your lighttpd ssl.conf:
    ssl.cipher-list = "RC4-SHA:AES256-SHA:AES128-SHA:DES-CBC3-SHA"
    ssl.honor-cipher-order = "enable"
    ssl.use-sslv2 = "disable"
    ssl.use-sslv3 = "disable"
    ssl.use-compression = "disable"

[/debian] permanent link


Offline Wikipeda on Mac OS X

Use WikiTaxi through Wine.

  1. Install Homebrew.
  2. brew install --use-gcc wine
  3. Prefix the WikiTaxi binaries with wine, e.g., wine WikiTaxi_Importer.exe, and otherwise follow the instructions on the WikiTaxi page.
It's not the prettiest solution, but it works quite well.

[/osx] permanent link


NFS doesn't work!

Unable to start nfsd or portmap in Debian? Getting weird errors in /var/log like the following?

The solution: make sure that the loopback device is up. E.g., make sure this is in /etc/network/interfaces:
auto lo
iface lo inet loopback

[/debian] permanent link


unpkg 4.5

It's released! Grab it while it's hot.

[/coding] permanent link


Compiling for older versions of Mac OS X

In preparation for a pending semi-major release of unpkg, I needed to compile some source for other architectures and older versions of Mac OS X on my Snow Leopard MacBook Pro. After quite some time DuckDuckGoing and wrestling with gcc, I came across these combinations of environment variables that seem to do the trick. (The following, of course, require the appropriate 10.4 and 10.5 SDKs to be installed in /Developer/SDKs.)

10.4 and PowerPC

CC="gcc-4.0 -arch ppc" \
CXX="g++-4.0 -arch ppc" \
CFLAGS="-mmacosx-version-min=10.4" \
-F/Developer/SDKs/MacOSX10.4u.sdk/System/Library/Frameworks \
-I/Developer/SDKs/MacOSX10.4u.sdk/usr/lib/gcc/powerpc-apple-darwin10/4.0.1/include \
-isystem /Developer/SDKs/MacOSX10.4u.sdk/usr/include" \
LDFLAGS="-arch ppc -mmacosx-version-min=10.4 \
-L/Developer/SDKs/MacOSX10.4u.sdk/usr/lib \
-F/Developer/SDKs/MacOSX10.4u.sdk/System/Library/Frameworks \
-Wl,-syslibroot,/Developer/SDKs/MacOSX10.4u.sdk" \
./configure --build=`uname -p`-apple-darwin --host=powerpc-apple-darwin
10.4 and i386
CC="gcc-4.0 -arch i386" \
CXX="g++-4.0 -arch i386" \
CFLAGS="-mmacosx-version-min=10.4" \
-F/Developer/SDKs/MacOSX10.4u.sdk/System/Library/Frameworks \
-I/Developer/SDKs/MacOSX10.4u.sdk/usr/lib/gcc/i686-apple-darwin10/4.0.1/include \
-isystem /Developer/SDKs/MacOSX10.4u.sdk/usr/include" \
LDFLAGS="-arch i386 -mmacosx-version-min=10.4 \
-L/Developer/SDKs/MacOSX10.4u.sdk/usr/lib \
-F/Developer/SDKs/MacOSX10.4u.sdk/System/Library/Frameworks \
-Wl,-syslibroot,/Developer/SDKs/MacOSX10.4u.sdk" \
10.5 and x86_64 -- the 10.4 gcc-4.0 can't compile for x86_64, so we have to use 10.5 instead.
CC="gcc-4.2 -arch x86_64" \
CXX="g++-4.2 -arch x86_64" \
CFLAGS="-mmacosx-version-min=10.5" \
-F/Developer/SDKs/MacOSX10.5.sdk/System/Library/Frameworks \
-I/Developer/SDKs/MacOSX10.5.sdk/usr/lib/gcc/i686-apple-darwin10/4.2.1/include \
-isystem /Developer/SDKs/MacOSX10.5.sdk/usr/include" \
LDFLAGS="-arch x86_64 -mmacosx-version-min=10.5 \
-L/Developer/SDKs/MacOSX10.5.sdk/usr/lib \
-F/Developer/SDKs/MacOSX10.5.sdk/System/Library/Frameworks \
-Wl,-syslibroot,/Developer/SDKs/MacOSX10.5.sdk" \
Then combine the binaries, like so: lipo foo-i386 foo-ppc foo-x86_64 -output foo-fat -create.

[/osx] permanent link


Converting a .mkv file to .mp4 to play on the iPad

Problem: you acquire a MKV file you want to play on your iPad. It's already H264/AAC, which the iPad can handle, so you don't want to re-encode it.

Solution: extract the tracks with mkvextract, and mux them into an iPad-friendly MP4 with MP4Box.

  1. Get a GPAC Subversion snapshot from http://gpac.sourceforge.net/.
  2. Build it. It's a bit finicky, and it will probably crap out looking for nonexistent wxwidgets libraries, but hopefully before that it'll create an MP4Box binary in bin/gcc.
  3. Grab a Mac OS X binary of MKVToolnix. Its dependencies are pretty extensive, so I used the binary available here.
  4. Determine what's in the mkv container: Mkvtoolnix.app/Contents/MacOS/mkvinfo Foo.mkv
  5. It'll tell you that, e.g., track one is H264 video at 24 fps, and track two is AAC.
  6. Extract the tracks accordingly: Mkvtoolnix.app/Contents/MacOS/mkvextract tracks Foo.mkv 1:Foo.h264 2:Foo.aac
  7. Put them back together in an MP4: MP4Box -add Foo.aac -add Foo.h264 -fps 24 Foo.mp4

Sometimes, though, you'll get audio in a non-AAC format (e.g., MP3 or AC3). This requires conversion.

  1. Download, make, and install the libfaac library. (We'd use FFmpeg's internal AAC encoder, but currently it's broken.)
  2. Do the same with FFmpeg, making sure to point it to libfaac appropriately.
  3. Convert the extracted audio file to AAC: ffmpeg -i Foo.ac3 -acodec libfaac -ab 192k Foo.aac
  4. Then use MP4Box as before, but with your new .aac file.
Thanks to this thread and this blog post.

[/coding] permanent link


DuckDuckGo search box in JavaScript

Sometimes, at work, we have terrible internet latency. This makes every page loaded count. I use DuckDuckGo as my primary search engine, and sometimes poor caching and latency combine to create a few seconds between starting to load the page and entering my query. So I whipped up a little Bookmarklet to bring up an equivalent search box, instantly. Hooray JavaScript! (The only time I'll ever say that...)

Here, drag: DDG

Works wonderfully as a hot-key, e.g., cmd-2 in Safari as the second link in my Bookmarks bar.

[/coding] permanent link


METAL Basic for the Mac (or: yet another young kid's introduction to programming)

Like many programmers, I was introduced to the craft through a BASIC dialect, specifically a MacOS-only one named METAL. I was taking extracurricular classes in C++ at the time, but the SIOUX (Simple Input Output User eXchange, i.e., the I/O offered by cin/cout) Mac runtime we used couldn't do anything graphical.

I can't remember how I stumbled upon METAL, but I have fond memories of heading to my friend's house, putting Cowgirl by Underworld on repeat, and seeing what we could get this simple language to do. We never created anything ground-breaking, of course -- a few simple point-'n'-shoot pseudo-3D games, for example -- but man would those coding sessions be serious.

Upon entering high school, we moved on to bigger and better languages. My friend found the Macromedia Flash MX (ca. 2002?) application on our school's computers, and I somehow dove deep into Java. I remember when version 1.5 (...5.0) came out, and thinking that generics and auto(un)boxing were the shit.

Anyway. METAL v1.7.3 was nevertheless my bread and butter for a few years. I recall checking its website once, noticing that an incredible v2.0 was on its way. To this day, the website still has "METAL v2.0 ULTRA" listed as ??? percent done.

That is, the Wayback Machine's version of the website. Naturally, the author graduated, and his personal website disappeared. It's still available here: http://web.archive.org/web/20080327004511/www.iit.edu/~sarimar/GDS/. METAL can be downloaded from there.

It still runs on Snow Leopard through Rosetta. Trying to run code results in a crashing runtime, but when "compiled" the resultant executable works fine.

If I ever stumble upon some old code I've written in METAL I'll post the atrocities. And I owe its author, Marin Saric, a beer or two.

[/coding] permanent link


What CPUs do the Amazon EC2 High-Memory On-Demand Instances use?

I can't vouch for the Extra Large or Double Extra Large ones, but the Quadruple Extra Large ones use Nehalem-based Gainestown Intel Xeon X5550 @ 2.67GHz.

processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 26
model name      : Intel(R) Xeon(R) CPU           X5550  @ 2.67GHz
stepping        : 5
cpu MHz         : 2666.760
cache size      : 8192 KB
physical id     : 0
siblings        : 1core id         : 0
cpu cores       : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 11
wp              : yes
flags           : fpu tsc msr pae mce cx8 apic mca cmov pat pse36
clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc
pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr dca popcnt lahf_lm
bogomips        : 5337.91
clflush size    : 64
cache_alignment : 64
address sizes   : 40 bits physical, 48 bits virtual
power management:

[/hpc] permanent link

The Authoritative List of Packages Needed to Comile the WikiReader Firmware

This is for a 32-bit Ubuntu Karmic schroot (as outlined in doc/Using-schroot.text) on a 64-bit Debian stable installation.


[/wikireader] permanent link


Using SheepShaver on Mac OS X

Build as listed on the website. Video uses X and audio goes through CoreAudio, so you don't need SDL. A few other bits of information, though:

[/osx] permanent link


Help! I have no /dev/dsp!

Just modprobe snd_pcm_oss and snd_mixer_oss, and restart udev. Then relive all your old Mac game meories through SheepShaver...

[/debian] permanent link


How to type accents in X

You'll want to use something called the Compose Key. Good instructions on this site, but here's a quick installation summary:

Mmm, háčeky.

[/debian] permanent link


My Terminal Emulator Setup in Debian

On my Debian boxes I use evilwm as my window manager, because it gets out of my way in just the way I like. Hence, I don't use gnome-terminal or kterm or ETerm or what have you; xterm has worked perfectly fine.

Nevertheless, I frequently spawn a lot of terminals, and the 5MB that each xterm process eats up (not including the 3MB bash does) add up after I fill a few virtual desktops with terminals. Xterm is also quite slow: cat'ing a 15K line banner output takes ~4 seconds on my netbook.

Hence, I went looking for replacements. Rxvt takes ~2MB per instance and cats the same file in ~1.2 seconds. Mrxvt takes ~3MB and ~0.5 seconds. Rxvt-unicode takes the cake, though, at ~3 MB and ~0.3 seconds. The real kicker, though, is its "daemon" mode, in which only once process is spawned for all "client" terminals, saving a lot of RAM. (Just hope it doesn't crash!)

Here's my setup:

  1. apt-get install rxvt-unicode-lite
  2. ~/.Xresources: (for a nice black-on-white xterm-like experience)
    URxvt*scrollBar: false
    URxvt*reverseVideo: true
  3. sudo update-alternatives --config x-terminal-emulator and select urxvtcd
Now I save 4-5MB of RAM for every terminal I have open, and when that gets to ~10 terminals that's not an insignificant amount of memory on a machine with 1GB of memory (and even more important on my lower-spec'd machines).

[/debian] permanent link


How to set up the QT4 WikiReader simulator in Debian unstable

  1. Certain build scripts require Python 2.6, so find some way to install it (Debian experimental, by source, etc.)
  2. apt-get install libqt4-dev qt4-qmake netpbm gforth build-essential php5-cli sqlite3
  3. git clone git://github.com/wikireader/wikireader.git
  4. cd wikireader/samo-lib/include
  5. Copy config.h-default to config.h and uncomment the SIMULATOR #define.
  6. make qt4-simulator
  7. Breakage will occur. Be merciless with hackage! I had to change a few lines of C and asm when functions weren't defined, etc.
  8. mkdir image work
  9. make DESTDIR=image WORKDIR=work XML_FILES=xml-file-samples/japanese_architects.xml index parse render combine (per doc/QuickStart)
  10. make DESTDIR=image install
  11. cp host-tools/qt4-simulator/bin/wikisim image/
  12. cd image
  13. ./wikisim
  14. It will segfault soon enough, but you'll hopefully be able to get in a bit of reading on Japanese architects. Good luck!

[/wikireader] permanent link


Installing SheepShaver and BasiliskII on Debian

I've tested SheepShaver with Mac OS 7.6.1 and Mac OS 8.1 with a 4MB OldWorld ROM. Sound works (even though ESD isn't used), video works, haven't tried ethernet. (all done on my current Debian unstable Intel Atom netbook.) The same instructions work for BasiliskII.

  1. Install the following packages: build-essential autoconf2.59 automake1.4 libxt-dev libxext-dev libxxf86dga-dev libxxf86vm-dev
  2. Follow the instructions here to download and the source from CVS, and make links as directed.
  3. Patch autogen.sh in src/Unix thusly:
    --- old/SheepShaver/src/Unix/autogen.sh 2007-06-13 14:09:05.000000000 +0200
    +++ ss2/SheepShaver/src/Unix/autogen.sh 2009-09-15 12:21:32.000000000 +0200
    @@ -40,13 +40,13 @@
     aclocalinclude="$ACLOCAL_FLAGS"; \
     (echo $_echo_n " + Running aclocal: $_echo_c"; \
    -    aclocal $aclocalinclude; \
    +    aclocal-1.4 $aclocalinclude; \
      echo "done.") && \
     (echo $_echo_n " + Running autoheader: $_echo_c"; \
    -    autoheader; \
    +    autoheader2.59; \
      echo "done.") && \
     (echo $_echo_n " + Running autoconf: $_echo_c"; \
    -    autoconf; \
    +    autoconf2.59; \
      echo "done.") 
     rm -f config.cache
  4. The output should look something like this:
    SheepShaver configuration summary:
    SDL support ...................... : none
    FBDev DGA support ................ : yes
    XFree86 DGA support .............. : yes
    XFree86 VidMode support .......... : yes
    Using PowerPC emulator ........... : yes
    Enable JIT compiler .............. : yes
    Enable video on SEGV signals ..... : yes
    ESD sound support ................ : no
    GTK user interface ............... : no
    mon debugger support ............. : no
    Addressing mode .................. : real
    Bad memory access recovery type .. : siginfo
    Configuration done. Now type "make".
  5. Patch sys_unix.cpp in src/Unix thusly:
    --- sys_unix.cpp.old    2009-09-15 12:28:54.000000000 +0200
    +++ sys_unix.cpp        2009-09-15 12:29:06.000000000 +0200
    @@ -883,11 +883,6 @@
    -               if (fh->cdrom_cap & CDC_DRIVE_STATUS) {
    -                       return ioctl(fh->fd, CDROM_DRIVE_STATUS, CDSL_CURRENT) == CDS_DISC_OK;
    -               }
                    cdrom_tochdr header;
                    return ioctl(fh->fd, CDROMREADTOCHDR, &header) == 0;
     #elif defined(__FreeBSD__) || defined(__NetBSD__)
  6. Make.
  7. If SheepShaver gets angry about ERROR: Cannot map Low Memory Globals: Permission denied. try: sudo sysctl -w vm.mmap_min_addr=0
To be continued...

[/debian] permanent link


Yet Another MSI Wind Hackintosh Tutorial

Just contributing my method. General outline: install OSX on an external drive from a MacBook, then copy the installation to the internal Wind drive.


  1. Boot the MacBook from the Leopard DVD.
  2. Partition the external drive, with a GUID scheme.
  3. Reboot and install the 10.5.7 update (newest 10.5.8 doesn't have good support at time of writing), reboot.
  4. Install CyberGreg's driver pack.
  5. Install Chameleon on the external drive.
  6. Boot the Wind from the external drive.
  7. Use SuperDuper to copy the OSX installation to the interal drive.
  8. Run Chameleon again on the interal drive.

[/osx] permanent link


When bit twiddling, make sure to use unsigned types.

This public service announcement was brought to you by the "Why The Fuck Don't My Bit Shifts Work" Department.

Goodness, that was amateur.

[/coding] permanent link


GMABooster segfaults on my machine!

Run it is root.

[/debian] permanent link


Benchmarking Intel's UXA, round 2

Debian unstable just upgraded to xserver-xorg-video-intel version 2.8.0. (released two days ago -- who says Debian is out of date?) This release is notable in that it removes support for EXA and DRI1, leaving only UXA and DRI2 for acceleration. Considering my previous benchmarking attempts, I thought it was only right to check the new drivers.

UXA on 2.8 is still slower than EXA on 2.7, but not by much. Notably, it's much faster than UXA on 2.7. I'm sure further releases will only lower these numbers further.

[/debian] permanent link


Installing Grub 2 on Debian sid

Sid has recently moved grub to grub-legacy and made grub 2 the default. My installation went fine -- all I had to do was remove grub-legacy (because of some conflicting man pages) and install os-prober and recreate grub.cfg to recognize my WinXP partition.

Hooray new packages in Debian!

[/debian] permanent link


Benchmarking Intel's UXA vs EXA

There's been a lot of development with regard to Intel's new UXA acceleration framework for its GMA integrated graphics chips. Benchmarks over at Phoronix have shown mixed results, so I decided to run some tests myself.

Tests run on an Asus Eee PC 900HA, with a 1.6GHz Atom N270 processor and integrated GMA 950 graphics, gtkperf 0.40, default Debian kernel 2.6.30, Xorg 1.6.2, and Intel driver 2.7.1. (i.e., an up to date unstable installation.)

So it looks like I'm sticking with EXA for now. If you have any tips as to boost either of these numbers, do let me know.

[/debian] permanent link


How to create a custom Debian package

I'm nowhere close to having the knowledge of a true Debian Developer, but I've used the OS enough to want to create my own packages. In this circumstance, I wanted a newer version of evilwm (it's 1.0.0 in the repo, and 1.0.1 is the newest release) and I wanted to change the mouse button behaviour. The steps I followed (I'm sure there's a cleaner way -- do inform me).

  1. apt-get install devscripts build-essential
  2. apt-get build-dep evilwm
  3. apt-get source evilwm
  4. Patch/hack away...
  5. debchange -n and document your non-maintainer changes
  6. debuild -us -uc so we don't try to sign it
  7. dpkg -i the resultant package

[/debian] permanent link


Getting an Elantech touchpad working in Debian squeeze (or: how to disable tap-to-click on an Eee 900HA)

Tap to click on a trackpad is a terrible idea. When I want to click, I'll hit the damn button myself, thank you very much. My ratio of unwanted to wanted tap-to-clicks is understandably undefined.

Three systems need to be configured for this to work: the kernel, the Xorg driver, and the X server itself.

  1. The kernel
    As documented here, recent kernels don't need to be patched; they just need CONFIG_MOUSE_PS2_ELANTECH enabled. Sadly, because of other issues the maintainers know about, it won't be enabled in mainline Debian kernels anytime soon. So either you'll have to build a kernel yourself, or just take the psmouse.ko I've created from 2.6.30, and pop it in the apropriate location in /lib/modules.
  2. The Xorg driver
    Debian testing ("squeeze" at time of writing) has an old version of xserver-xorg-input-synaptics that doesnt have the necessary Elantech patches. Unstable does, but pinning the binary has other dependencies that would pretty much require upgrading all of X, which I'd rather not do. So you can either pin the source file and build it yourself, or download and install the package I've made. Grab it here.
  3. The X server
    We just need to poke aorund xorg.conf a bit. What worked for me:
    Section "InputDevice"
            Identifier      "Configured Mouse"
    	Option		"CorePointer"
            Driver          "synaptics"
            Option          "Device"                "/dev/input/event10"
            Option          "Protocol"              "auto-dev"
            Option          "SHMConfig"             "true"
    	Option		"MaxTapTime"		"0"
            Option          "VertTwoFingerScroll"   "1"
            Option          "HorizTwoFingerScroll"  "1"
Good luck!

[/debian] permanent link

Running Debian testing on an Eee PC 900HA

Great news: almost everything works out of the box. I was concerned about hardware support, and struggling with wireless, so I just apt-pinned the latest kernel from unstable, and it works perfectly.

  1. X11 at 1024x600, DRI
  2. Sound
  3. Ethernet
  4. Touchpad w/ two-finger scrolling (although with annoying-as-hell tap to click -- more on that later)
  5. Wireless
  6. Camera (w/ Skype)
  7. Suspend to RAM
Protip: make sure to modprobe acpi-cpufreq for frequency scaling.

I'm using evilwm as my window manager, because I love how it gets out of my way and lets me do my work. I'd post a screenshot, but there's no eye-candy to show off... =) I've also had to hack Firefox into shape so it doesn't take up silly amounts of vertical pixels.

[/debian] permanent link


How to Install Debian on an Eee PC 900HA

I just picked up an Asus Eee PC 900HA. It, currently, has the best mix of size, cost, and capabilities for what I want in a netbook. I'm keeping Windows on it to use Ableton Live and for "just in case" purposes. Here's documentation of what I've done to re-partition, keeping the provided Windows installation, and install Debian testing.

  1. Download and repartition using a GParted bootable USB drive. I used the Windows-based choice 3.
  2. Make sure to set the USB disk as higher priority than the internal drive, or else it'll boot from the HD every time.
  3. My drive came partitioned into four partitions:
    1. Windows boot (~80 GB)
    2. Windows free (empty, ~60 GB)
    3. Windows restore partition (~8 GB)
    4. Extra restore info? (~40 MB)
    You should be safe deleting all but the first partition and going from there, but it's not as if I'm wanting for space on this netbook, so I let 'em stay.
  4. Follow the instructions on this page of the Debian installation manual.
To be continued...

[/debian] permanent link


Does Ableton Live work on an Eee PC 900HA?

Very much so. With two decks, equalizers, and a few other affects, CPU usage doesn't appear to go higher than ~20% or so. This may just become my new gig rig!

Just make sure to install QuickTime if MP3s are at all part of your setup. Cheers.

[/music] permanent link

How to configure and make gcc with GRAPHITE

Done on a standard Debian installation with a sane toolchain installed.

  1. Download and untar libgmp, configure with ./configure --enable-cxx, and install.
  2. Download, untar, patch (according to their website), configure, and install libmpfr.
  3. You know the drill by now: ppl.
  4. apt-get install automake1.7 (for cloog-ppl)
  5. Who said we were anywhere close to done? cloog-ppl. Its configure script is really picky, though; I had to explicitly use --with-gmp=dir and --with-ppl=dir.
  6. Grab gcc-4.4.0, untar, and configure. These are the options I used (following Debian's gcc -v lead): $ ../gcc-4.4.0/configure --enable-languages=c,c++,fortran --prefix=/home/timdoug/local/ --enable-shared --with-system-zlib --without-included-gettext --enable-threads=posix --with-gmp=/home/timdoug/local/ --with-mpfr=/home/timdoug/local/ --enable-mpfr --with-ppl=/home/timdoug/local/ --with-cloog=/home/timdoug/local/ (note the different object directory!)
  7. The configure script should spit out something like this, amongst its output:
    checking for correct version of gmp.h... yes
    checking for correct version of mpfr.h... yes
    checking for version 0.10 of PPL... yes
    checking for correct version of CLooG... yes
  8. make, and grab a sandwich.
  9. Curse profusely when it doesn't compile.
To be continued...

[/gcc] permanent link


Installing Debian on a PlayStation 3 and running a Cell/SPU "Hello World"

What could be better than flexing a bit of matrix-multiply muscle? We're going to use Daniel Hackenberg's hand-coded asm routines, because they achieve ridiculous performance.

  1. Follow these instructions.
  2. Install some packages: apt-get install build-essential rpm libc6-dev-ppc64 gcc-4.3-spu spu-tools
  3. Grab the following RPMs off of CellSDK-Devel-Fedora_3.
    1. ps3:/# rpm2cpio /home/timdoug/libs/cell-libs-3.0-16.ppc.rpm | cpio -id
    2. ps3:/# rpm2cpio /home/timdoug/libs/cell-libs-3.0-16.ppc64.rpm | cpio -id
    3. ps3:/# rpm2cpio /home/timdoug/libs/cell-libs-devel-3.0-16.ppc.rpm | cpio -id
    4. ps3:/# rpm2cpio /home/timdoug/libs/cell-libs-devel-3.0-16.ppc64.rpm | cpio -id
  4. Debian ships with a 32 bit libspe2, but we need a 64 bit version:
    1. wget http://ftp.debian.org/debian/pool/main/libs/libspe2/libspe2_2.2.80-95.orig.tar.gz
    2. Add -m64 to CFLAGS in make.defines.
    3. make
  5. Add spufs /spu spufs defaults 0 0 to /etc/fstab.
  6. mkdir /spe && mount /spe
  7. wget "http://www.tu-dresden.de/hpcadm/dcount/dcount.php?package=matmul&get=matmul.tar.gz"
  8. Untar matmu.tar.gz and patch along these lines:
    diff -urN matmul-old/COMPILE.sh matmul/COMPILE.sh
    --- matmul-old/COMPILE.sh	2008-02-29 06:30:08.000000000 -0500
    +++ matmul/COMPILE.sh	2009-06-06 17:12:39.000000000 -0400
    @@ -29,15 +29,15 @@
     ${CELL_BIN}/spu-gcc -o matmul_spu matmul_spu.o matmul_spu_simd.o
     # embedd SPE object file into PPE object
    -echo "${CELL_BIN}/ppu-embedspu -m64 matmul_spu matmul_spu matmul_spu-embed64.o"
    -${CELL_BIN}/ppu-embedspu -m64 matmul_spu matmul_spu matmul_spu-embed64.o
    +echo "${CELL_BIN}/embedspu -m64 matmul_spu matmul_spu matmul_spu-embed64.o"
    +${CELL_BIN}/embedspu -m64 matmul_spu matmul_spu matmul_spu-embed64.o
     # compile PPE code
    -echo "${CELL_BIN}/ppu-gcc -W -Wall -O3 ${INC_PPU} -c matmul_ppu.c"
    -${CELL_BIN}/ppu-gcc -W -Wall -O3 ${INC_PPU} -c matmul_ppu.c
    +echo "${CELL_BIN}/gcc -m64 -W -Wall -O3 ${INC_PPU} -c matmul_ppu.c"
    +${CELL_BIN}/gcc -m64 -W -Wall -O3 ${INC_PPU} -c matmul_ppu.c
     # link SPE and PPE object files together
    -echo "${CELL_BIN}/ppu-gcc -o matmul matmul_ppu.o matmul_spu-embed64.o -lspe2"
    -${CELL_BIN}/ppu-gcc -o matmul matmul_ppu.o matmul_spu-embed64.o -lspe2
    +echo "${CELL_BIN}/gcc -m64 -o matmul matmul_ppu.o matmul_spu-embed64.o -lspe2 -lpthread -L/home/timdoug/libspe2-2.2.80-95"
    +${CELL_BIN}/gcc -m64 -o matmul matmul_ppu.o matmul_spu-embed64.o -lspe2 -lpthread -L/home/timdoug/libspe2-2.2.80-95
     rm -f matmul_spu *.o
  9. Open up spu-top in another terminal to check up on things.
  10. ./matmul -s 6 -m 3072
timdoug@ps3:~/matmul$ ./matmul -s 6 -m 3072

Fast matrix multiplications on Cell (SMP) systems.
Copyright (C) 2007  Daniel Hackenberg, ZIH, TU-Dresden

Running matrix multiplication of 3072x3072 matrices using 6 SPEs...
Initializing arrays with random numbers... done!
Starting SPE calculations...

Performance results:
Performance of SPE  0: 25.35 GFLOPS
Performance of SPE  1: 25.35 GFLOPS
Performance of SPE  2: 25.36 GFLOPS
Performance of SPE  3: 25.36 GFLOPS
Performance of SPE  4: 25.36 GFLOPS
Performance of SPE  5: 25.36 GFLOPS
Aggregated performance for all 6 SPEs: 152.14 GFLOPS.

PPE-measured performance of matrix multiplication using 6 SPEs: 152.11 GFLOPS.
(of 153.60 GFLOPS theoretical peak at 3200 MHz clock frequency)
152 SP GFLOPS is crazy absurd for $400. The above taks ~7.5 seconds to run, but most of the time is in initializing the matrix with random values!

[/ps3] permanent link


Help! I'm running out of stack space!

Edit /etc/security/limits.conf, open up a new shell, and have fun with ulimit.

[/debian] permanent link

Fun summer projects

So. With a summer ahead of me, there are a few things I want to get into now that I have the time. A list:

[/general] permanent link

Reason N why I love Python

Code like this: ",".join(nodes_list[:num_nodes])
Objects, strings (finally sane in 3.0), lists, list slices...

[/python] permanent link


Reseting the root passwd on a PS3 Debian installation

I conveniently forgot the password for my year-old PS3 installation, but conveniently recognized that kboot is a little Linux installation in and of itself. Hence, mounting /dev/ps3da1, chrooting into it, and running passwd does the trick.

[/ps3] permanent link


How to install rsh on Debian

Debian is aptly (hah!) fickle with regard to rsh -- by default, if you have OpenSSH installed, rsh is just a symlink to it. For my research I need rsh in order to get around the overhead that ssh incurs; security isn't an issue because the machines are on a private network anyway.

  1. apt-get install rsh-redone-client rsh-redone-server
  2. Add appropriate hostnames to ~/.rhosts
  3. Edit /etc/hosts.deny and add in.rshd in.rlogind: ALL
  4. Edit /etc/hosts.allow and add in.rshd in.rlogind: [hostnames] as necessary.
Have (insecure) fun!

[/debian] permanent link

High-Performance Computing and the PlayStation 3

I purchased a PS3 last summer with the express intent of running some simulations and benchmarks (and playing Gran Turismo 5 Prologue), but never got around to it (the first part, that is).

A few interesting links and PDFs:

I'm intrigued by CUDA as well. We'll see where that goes.

[/hpc] permanent link

How to Install and Run HPL with GotoBLAS on Linux

GFLOPs are fun. Here's how to determine your cluster's performance.

  1. (make sure you have a proper build system: gcc, make, etc. Debian: apt-get install build-essential)
  2. Install OpenMPI and gfortran. On Debian, it's as simple as apt-get install gfortran libopenmpi-dev openmpi-bin.
  3. Download and compile GotoBLAS. It's generally lauded as the fastest BLAS for (at least) x86 machines. Compilation is a simple ./quickbuild.64bit (or 32, as appropriate) in the root directory of the tarball.
  4. Download and untar HPL. I used 1.0a -- 2.0 wouldn't compile for me.
  5. Create a Make.[arch] in the root dir of the hpl folder, and configure accordingly. Appripriate example diff against setup/Make.Linux_PII_CBLAS:
    --- setup/Make.Linux_PII_CBLAS	2004-01-22 00:13:11.000000000 -0500
    +++ Make.timdoug	2009-06-01 00:23:29.000000000 -0400
    @@ -61,7 +61,7 @@
     # - Platform identifier ------------------------------------------------
     # ----------------------------------------------------------------------
    -ARCH         = Linux_PII_CBLAS
    +ARCH         = timdoug
     # ----------------------------------------------------------------------
     # - HPL Directory Structure / HPL library ------------------------------
    @@ -81,9 +81,9 @@
     # header files,  MPlib  is defined  to be the name of  the library to be
     # used. The variable MPdir is only used for defining MPinc and MPlib.
    -MPdir        = /usr/local/mpi
    -MPinc        = -I$(MPdir)/include
    -MPlib        = $(MPdir)/lib/libmpich.a
    +MPdir        = /usr
    +MPinc        = -I$(MPdir)/include/mpi
    +MPlib        = -lmpi
     # ----------------------------------------------------------------------
     # - Linear Algebra library (BLAS or VSIPL) -----------------------------
    @@ -92,9 +92,9 @@
     # header files,  LAlib  is defined  to be the name of  the library to be
     # used. The variable LAdir is only used for defining LAinc and LAlib.
    -LAdir        = $(HOME)/netlib/ARCHIVES/Linux_PII
    +LAdir        =
     LAinc        =
    -LAlib        = $(LAdir)/libcblas.a $(LAdir)/libatlas.a
    +LAlib        = ~/GotoBLAS/libgoto.a
     # ----------------------------------------------------------------------
     # - F77 / C interface --------------------------------------------------
    @@ -156,7 +156,7 @@
     #    *) call the BLAS Fortran 77 interface,
     #    *) not display detailed timing information.
    +HPL_OPTS     = 
     # ----------------------------------------------------------------------
    @@ -173,7 +173,7 @@
     # On some platforms,  it is necessary  to use the Fortran linker to find
     # the Fortran internals used in the BLAS library.
    -LINKER       = /usr/bin/g77
    +LINKER       = /usr/bin/gcc
     ARCHIVER     = ar
  6. In hpl/bin/[arch], tweak HPL.dat. Example file:
    HPLinpack benchmark input file
    Innovative Computing Laboratory, University of Tennessee
    HPL.out      output file name (if any)
    6            device out (6=stdout,7=stderr,file)
    1            # of problems sizes (N)
    16384        Ns
    1            # of NBs
    128          NBs
    0            PMAP process mapping (0=Row-,1=Column-major)
    1            # of process grids (P x Q)
    1            Ps
    2            Qs
    16.0         threshold
    1            # of panel fact
    2            PFACTs (0=left, 1=Crout, 2=Right)
    1            # of recursive stopping criterium
    4            NBMINs (>= 1)
    1            # of panels in recursion
    2            NDIVs
    1            # of recursive panel fact.
    2            RFACTs (0=left, 1=Crout, 2=Right)
    1            # of broadcast
    1            BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)
    1            # of lookahead depth
    0            DEPTHs (>=0)
    0            SWAP (0=bin-exch,1=long,2=mix)
    64           swapping threshold
    0            L1 in (0=transposed,1=no-transposed) form
    0            U  in (0=transposed,1=no-transposed) form
    1            Equilibration (0=no,1=yes)
    8            memory alignment in double (> 0)
    This is appropriate for a dual-core, 4GB RAM system. Important values to change:
    • N -- problem size. Start at ~1000 and ramp up until you hit the limit of your RAM. Note: it's quadratic (dimension of the matrices).
    • NB -- block size. 128 works well for me. Others suggest 80, 160, or 256. Experiment.
    • P and Q -- these multiplied should be the number of cores in your cluster. Certain configurations work better than others; do test.
    HPL.dat tweaking is a kind of black magic -- check the tubes for further information.
  7. Run GOTO_NUM_THREADS=1 mpiexec -np [num processes] ./xhpl
Using these instructions, I achieved 5.5 GFLOPs per core and 10 GFLOPs in total on a Pentium D 930 machine, 9.5 GFLOPs per core and 18 GFLOPs on a Core 2 Duo E6700 processor, and 50.6 GFLOPs on an Amazon EC2 "High-CPU Extra Large Instance" (dual quad-core Xeon E5345s):
domU:~/hpl/bin/timdoug# GOTO_NUM_THREADS=1 mpiexec -np 8 ./xhpl
HPLinpack 1.0a  --  High-Performance Linpack benchmark  --   January 20, 2004
Written by A. Petitet and R. Clint Whaley,  Innovative Computing Labs.,  UTK
T/V                N    NB     P     Q               Time             Gflops
WR01L2C4       20000   128     2     4             105.33          5.064e+01
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0067492 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0065323 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0012449 ...... PASSED

[/hpc] permanent link

© 2006-14 timdoug | email: "me" at this domain
So necessary