ANU Film Group 2005 Semester 1 Calendar

The ANU Film Group resumed showing films this week. Here is an ICal calendar of the program that I whipped up today.

V1280 photos

In reference to my earlier message regarding the death of a CPU board on our Sun V1280, here are some photos. (all photos here)

Here's the new board, sans RAM:	Here's the old board, with RAM:
Insertion of the new board, demonstrating the anti-gravity device:	A close-up of one of the CPUs in the dead board:

A board has died, long live the new board!

Our group has a Sun V1280 entry-level midrange server, for computational chemistry, software development and experiments.

It has 12 900 MHz UltraSPARC III Cu processors and 24 GB of RAM (3 CPU boards with 4 CPUs and 8 GB each). It's a pretty cool machine.

Last Friday night, it rebooted without warning (and without crashing). It seems that something went wrong with one of the CPU boards, so the system controller forced a reset. It came up with only 8 CPUs and 16 GB: one of the CPU boards failed POST.

Thankfully we have Sun support, so they (eventually) diagnosed the problem and shipped us a replacement board (along with a replacement power supply for one whose fan was making funny noises).

I was hoping to be able to utilise the V1280's (and Solaris') hot-swappability when replacing the CPU board. Unfortunately the dead board was so broken the system controller (lights-out-manager, which appears to be an UltraSPARC IIe) couldn't turn the power off to the dead board only. So I had to bring the whole machine down in order to replace it.

The replacement system board had 4 900MHz CPUs installed (the CPUs aren't user-serviceable components), but no RAM, so I had to move 32 sticks (256 MB each) of RAM from the dead board to the new board. Ouch thumbs.

Pictures to follow

MTB Ride: Black Mountain to Mount Stromlo (and back)

Yesterday, Victoria and I did a nice ride from Black Mountain to Mount Stromlo.

We followed part of leg 5 of the Triple Triathlon, starting from Rani Road (soon to be demolished and covered over by the great Gungahlin Drive Extension (GDE)) and over Dairy Farmer's Hill, down to Coppins Crossing, up over Bluett's (the impossible climb, as far as I am concerned!), and up the back of Mt Stromlo to the top.

There we found that the cafe is now re-opened: Michael's of Stromlo (website is currently empty).

(The Red-Belly Black Cafe used to have a cafe on top of Mt Stromlo, but it was devastated by the 2003 bush-fires.)

After riding for a few hours in fairly warm conditions, we decided that lunch at the top of Mt Stromlo was ideal. We could then cruise back to Black Mountain along the bike path.

Michael's isn't up to the standard of the Red-Belly Black, but it is cheaper. We ordered two "Hamburgers with Pineapple, Bacon and Cheese". The lady at the counter said, "do you want egg with that?", to which we replied, "yes, please!". Some twenty minutes later the burgers arrived (there was a warning in the menu saying to expect delays in delivery, which I had thought at the time was weird). Delivered as just an open burger with a good solid bun and plenty of crispy bacon, beetroot, pineapple, cheese, lettuce, and a good patty. I put the two parts together and devoured, although I did have some containment issues towards the end. Yum. Not bad for $5 each, with a fantastic view (the menu also states, "the cafe with the million-dollar view").

If I recall correctly, Michael's is now open Wednesday to Sunday, 10am - 5pm. A good place for a pit-stop.

After finishing lunch, we hurtled down the Mt Stromlo road. I hadn't ridden up or down the main road since the fires (I did leg 5 of the triple tri last year, but that goes down fire trails so the view is a bit different since the condition of the trails is so bad that you're looking at the ground all the time). I was surprised at the view you get going down the road now, since all the trees have been either burnt or chopped down. It's sad, but the view is incredible.

An uneventful ride home along the bike path. Despite being tired it took us less than half the time of the ride out. I think I need to do some more riding before I do all 3 legs of the Triple Quad on March 20.

Pseudo-terminals and Xterms

Ok, so I've been trying to work out a nice way to get input and output from our simulator separate to the input and output to the simulated program. This is particularly important if we want to have user control of the simulator while the simulation is running.

This is exactly what pseudo-terminals are good for. On Solaris, see ptm(7D) and pts(7D) (or go to Sun's online web-pages: ptm(7D).

Essentially a pseudo-terminal is like a bidirectional pipe(2), but also with terminal semantics, so there are special characters so the user can do things like backspace, line erase, etc.

A pseudo-terminal has a master end and a slave end. The master end is where the user is (expecting output and reading user input), the slave end is where the program is (expecting user input and writing output). The slave end will respond to the usual terminal ioctl(2)'s that the program wants to use (for example, TCGETS, in order to get a copy of the termios structure from the terminal). It turns out (see below) that this ordering (master=user, slave=program) is crucial.

Ok, so I could use a pseudo-terminal, but I still needed a way for the user to input into that pseudo-terminal: I wanted to use the actual terminal for simulator input and control. It turns out that xterm(1) has a switch exactly for this purpose: -Sccn. The manpage says:

     -Sccn   This option specifies the last two  letters  of  the
           name  of a pseudoterminal to use in slave mode, plus
           the number of the inherited  file  descriptor.   The
           option  is  parsed ``%c%c%d''.  This allows xterm to
           be used as an input and output channel for an exist-
           ing  program  and  is  sometimes used in specialized
           applications.

However, modern pseudo-terminals are created using the /dev/ptmx and ptsname(3C) interface (rather than using the old, BSD-style search through /dev/ptyXY where X is [pqrs...] and Y is [0-9a-f]. The -Sccn switch was designed for ye-olde BSD type of pseudo-terminal days.

It turns out that newer versions of xterm know about this, and have extended the switch to allow -Sccc.../d where the ccc...'s are the trailing part of the slave device name (the basename) and the d is the device descriptor (with a / between them to separate them). However, Solaris's xterm doesn't support that.

But it also turns out that xterm ignores the cc characters these days anyway. How did I find this out? I read the source. Yuck, but true. Although I didn't read the source for the Solaris version of xterm it seemed likely that it ignored it too. Especially since Solaris's CDE terminal dtterm(1) also had a similar -Sccn switch and an extension switch:

        -Sc.n
            Equivalent to -Sccn, but provided for systems with
            a larger pseudo-terminal device name space.  The c
            argument specifies the last component of the
            pseudo-terminal device slave name.  The terminal
            emulator ignores this value and the value may be
            empty.  The n argument specifies the number of the
            file descriptor that corresponds to the pseudo-
            terminal device's already opened master side.

So I wrote a little program to test this out. See pseudo.c.

When I first tried this I was getting really weird results: the xterm wouldn't show anything the user typed, but the characters would flow one-by-one to the parent program (i.e., not line-based as I expected). Also, anything I wrote back from the parent program would not only get printed on the xterm window, but would also end up back in the input of the parent program's end of the pseudo-terminal.

After much fiddling and thinking, I realised that what I wanted to do with the pseudo-terminal is different to most examples of it: I wanted to give the master to the child program, not the slave: I wanted the child to be the user input and the parent to simply be a proxy between the user and the simulated program (going back to the original simulation IO problem). So after simply switching slave and master, the program worked as expected.

However, I also got an additional line of input (both appearing in the xterm and in the parent program's pseudo-terminal input). This line contained a hexadecimal number. After reading the xterm source, it turns out it writes the X-window id of itself to the terminal so the parent program can control it if it wants to. That also appears in the xterm itself since by default user input is echoed (terminal control local flag ECHO).

I'm not interested in the xterm's X-window id, and I figure it could be confusing to see an xterm pop up with 0xc0002 or somesuch in it. So there's some fudging going on that ensures that we don't see it: Firstly, before forking the child I get the pseudo-terminal (default) settings, and then I clear ECHO. So when the xterm pops up, it will write its X-window id, but it won't appear in the xterm. Then post-fork, the parent reads and discards a line from the xterm: this should be the xterm's window id. The parent then restores the original terminal settings, so ECHO is set if it was already, which appears to be the default.

To get this working I did a lot of reading: Stevens' "Advanced Programming in the UNIX Environment", the Solaris man-pages, and the sources to xterm (CVS) and expect (CVS).

Chainfire, by Terry Goodkind

Book 9 of the Sword of Truth

This is the first book of a trilogy that will complete the Sword of Truth series. Firstly, I have to say straight up, that the first book of the Sword of Truth, "Wizard's First Rule" is my favourite book. If you haven't read it, beg, borrow or buy it, and read it.

The rest of the series hasn't quite been up to the standard of Wizard's First Rule, but they've been pretty damn good reads nonetheless. While the first book was fairly standalone, the rest have clearly been part of an overall series. They have all been leading to a final confrontation between Richard Rahl and the Imperial Order (Emperor Jagang, the dreamwalker).

This book starts with Kahlan, Richard's wife and a major part of the series, going missing. And no-one but Richard remembers her. (that much is on the blurb on the back, no spoilers there, however: spoiler may be below!)

I found this extremely readable, as usual for Terry Goodkind's novels. While I occasionally got annoyed with his long soliloquies talking about the philosophy of freedom (somewhat preachy and repetitive) the story is intriguing. I particularly enjoyed the linking back to the first book with the Boxes of Orden, and since this will clearly be a big part of the next two books I think that will make a nice wrap-up to the series.

Highly recommended.

Pandora's Star, by Peter F. Hamilton

Book 1 of the Commonwealth Saga.

Finished this last week. Very good read, excellent world-building, with an interesting plotline and (mostly) believable vision of the future. The ending is intriguing, leaving you waiting for the next one (Judas Unleashed, due October 2005).

Highly recommended.

migraines suck

my first blog post.

had a migraine last night, after playing two games of ultimate in fairly stifling heat. i think this is the first migraine i've had in maybe 5 years. migraines suck big time.

i know when they're coming around because i start not seeing things: that is, there are gaps (white splotches) in my vision. all i can do then is go lie down somewhere quiet and close my eyes and hope i can get to sleep.

thankfully i got to sleep through most of it.

still had a headache all day today, but i came in to work for a few hours because it was andrew's mid-term seminar, and i'm on his supervisory panel. it went well, as expected.

so i decided to be a sheep and do a blog. yeah, that follows.