|
| Greg's diary |
| November 2002 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Friday, 1 November 2002 | Echunga | |
| Top of page | ||
| next day | ||
| last day |
Up early today and to the Wiredup expo, borrowing Yana's laptop, since I still don't have the spare parts for my laptop. Dell certainly doesn't get any recommendation from me.
At the expo, things were relatively slow. It also shows that we didn't have much idea of what we were up against; we still don't even have information sheets to hand out explaining who we are, and why. We were supposed to have access to somebody else's wireless network, but by the time I left, they still hadn't finished it. Did some talking to Dan Shearer, who was particularly active, but somehow managed to find time to introduce me to some people whom he thought worth knowing.
Back home and on with the book, moving on to the horrible theme of upgrading. Things are moving on a little, anyway.
Broken web site of the month: Wine Jobs Australia. They insist on Microsoft Internet Explorer, and won't give you access for anything else. Sometimes I wonder whether people run web sites to provide information, or to exclude people from their sites.
| Saturday, 2 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Yesterday was tiring, so up very late today, and didn't do too much. Updating old systems is a real pain. On the one hand, it's difficult to write about, and on the other hand it's clear that I'm going to have to do something more about the problem than documenting it, and that means that what I'm writing now will probably be obsolete even before it hits the press. I can't win.
| Sunday, 3 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More work on the book today. I had another book to review, and it made me look at the thing from another side, so I ended up redoing half of the Apache description.
In the afternoon, cooking with Yana. I had been looking for a Chinese claypot chicken recipe, and discovered to my surprise that there was none in any of my cookbooks. Found a couple on the web, of typical web quality, and managed to interpret one relatively successfully.
| Monday, 4 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Got a message from Berny Goodheart this morning; he sent it to auug-announce@auug.org.au, looking for the president. He doesn't seem to know that that's me. When replying, I managed to stuff up right royally and copy the entire membership list. It'll take me a while to live that one down. I obviously shouldn't be allowed to post without authorization.
More work on the book, and finally got the updating chapter finished. That's a load off my mind. Also spent some time tidying up my HTML; a lot seems to have changed since I learnt it.
Finally mowed the hay. For once, over the Melbourne cup week the weather looked acceptable. Yvonne is on tenterhooks, though.
| Tuesday, 5 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Somehow everything's happening at once again. Apart from the book, with which I made some progress, I also had to plan my trip to NOIE in Canberra, not made easier by the fact that Dell still haven't delivered my replacement hinge for my laptop, though I did get a message apologizing for the delay and explaining that it had taken 10 days to get the part number. As a result, had to move a lot of data to Yana's laptop, much to her pleasure. In the process, got interrupted by all sorts of other things; it looks like AUUG is going to be active next year.
| Wednesday, 6 November 2002 | Echunga → Canberra | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
To the airport this morning, feeling strangely tired, and didn't really wake up until I got to Melbourne. In Canberra, Hugh was there to pick me up, and back to IBM, where nothing seems to have changed much. Spent the afternoon trying to get firefly, Yana's laptop, to run two concurrent X servers. That worked, sort of, but the mouse went crazy. I can't make up my mind whether the dual server config confused it, or (more likely, I think) the APM, which never seems to have worked properly.
Back to Hugh's place in the evening, talking about old times. Pretty tired, early to bed.
| Thursday, 7 November 2002 | Canberra → Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
In to town with Hugh in the morning to await the arrival of Gordon and Con. I was right to have come the previous day: Con somehow had a messup with his ticket, and as a result he had to wait for another two flights before they got him there. As a result, didn't have much time to prepare, but that proved not to be a significant issue, since they didn't want to see our presentation anyway.
The upshot of the meeting was that we (AUUG) will have a significant slot in the meeting they're planning for mid-February. The NOIE people are surprisingly knowledgeable, thus the lack of a need to watch our presentation. Instead, we spent a lot of time discussing the issues of Open Source that interest the people who count in government. This looks like being fun.
After that, more deliberations, and off to the airport: our planes were all pretty much at the same time, and the Qantas Club was even fuller than usual for Canberra. Back home uneventfully. The weather has really picked up here, and fortunately it hasn't rained, nor does it seem to be about to. Looks like we'll get a good harvest of hay, which is particularly important this year: it's been really dry, and the hay harvest will be pretty lean.
| Friday, 8 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
In catch-up mode again today. Although I got my mail pretty much under control on the flight home last night (in fact, better than I've been for a while), inconsequentialities kept me busy most of the day. When I did get round to doing something on the book, discovered that Kernel ppp no longer worked on zaphod. We may be close to releasing FreeBSD 5.0, but it's going to be a bit of a shambles when it happens.
Postponed the ppp chapter and moved on to Vinum, where I discovered a new bug in pic which made it impossible to continue on that front. Frustrating day.
| Saturday, 9 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Yvonne is spending a long weekend participating in a horse training “clinic” to train Scarlett to load onto a float, a thing that has caused a surprising amount of difficulty. It's at Tilling Hill, about 7 km away on the delightfully named Rubbish Dump Road. What a name. It's round the old Jupiter Creek gold mines, so you'd think that Gold Mine Road would be a valid alternative.
Obviously we couldn't transport Scarlett there, so we rode over there and floated Darah back. Relatively uneventful, but it was a warm day, and we were pretty well finished when we got there.
Took it easy and didn't do much work.
| Sunday, 10 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More running away from work today. Instead, tried installing Wine on wantadilla. That didn't work well. First, it claimed that it needed the option USER_LDT in the kernel configuration file, but it proves that that's no longer the case in FreeBSD-CURRENT. But it complained greatly about the lack of a CD-ROM drive and sound hardware. Since my intention is to read the Oxford English Dictionary, which comes on a stupid non-standards-compliant CD-ROM (some sectors are not legible; they've put more effort into the copy protection than they have into the presentation), I decided I'd need a real CD-ROM drive, so started installing on echunga, where the installation didn't complete. On closer examination, it looks as if the port doesn't install all the programs. When I tried to run one, it complained about the lack of USER_LDT. This is a 4-STABLE machine, so decided to upgrade to the latest version and build a new kernel.
In the meantime, did some cooking, including making a batch of kimchi, in the process taking some photos in order to complete the recipe I wrote a while back.
After that, off to see how Yvonne and Scarlett were getting on with the float loading. It's a lot better than yesterday.
Brian Schulz started baling hay today, but it didn't take him long before he had another breakdown (he already had one on Monday when mowing the hay). He got it fixed, came back, and pretty soon after that had another one, so he only got about half of it baled today. We're running out of time: bad weather is forecast for Tuesday.
| Monday, 11 November 2002 | Echunga | Images for 11 November 2002 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More work on the book, and got two more chapters into more-or-less printable condition. Hopefully I can get that out of my hair soon.
Spent more time working on the Wine issue. Installed the new kernel on echunga, not without another run-in with the upgrade process. This really needs to be fixed. Wine itself wasn't much better: it's unnecessarily and confusingly verbose. For example, when looking for DLLs (dynamically linked libraries) it reports as an error every one that it doesn't find in a particular place, even though it's found later in a different place. Managed to get a solitaire game up, but when trying to start the Oxford English Dictionary software, it claimed that the license manager couldn't do its thing. Tried installing again, but the installer kept crashing, and debug attempts didn't work. I suspect that it would take a lot of work to get that fixed. I suppose VMware is the next thing to try.
Brian Schulz was due back today to bale hay, but by 2 pm he hadn't shown up. Apparently it was too hot to bale. I'm not clear whether he meant too hot for himself or for the hay, but the result was that he didn't show until 6 pm, by which time it was cooling down. He finished baling, but he wasn't able to get the stuff into the shed. And that with the threat of bad weather tomorrow. sigh.
| Tuesday, 12 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Catastrophe! It started raining at 5 am this morning, with our 400 bales of hay still outside in the paddocks. Got up in a foul mood, but things gradually calmed down. It'll be at least a few days now before we can put the hay in the sheds.
Got through my mail pretty quickly and spent some more time working on the book. I now have the networking section (about half the book) in review draft state, but I'm getting surprisingly little review feedback. Worked on the installation chapter and reorganized it completely. When I first wrote it, CD-ROMs were by no means universal, and a lot of machines couldn't boot them. Nowadays it's about the only way to install, so I've moved all the discussion of the other methods into the sidelines.
| Wednesday, 13 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Finally things are looking up with the book, so didn't do any work on it today beyond putting up the latest chapters on the web page. Instead, spent some time tidying up my office, which was sorely in need of it:
In the process found things on the big table to my left which had been piled up for at least 15 months, not to mention an old Tandem expense report dated 18 October 1985, not submitted. I was living in Germany at the time and had spent 6 weeks in Australia working for Tandem Australia. I assume that the expenses were paid without the German formalities; the sum was quite significant.
Guy Schubert here today, finally overhauling the sprinkler system. Not too early, either: it's getting warm again. Yesterday's rain was just to annoy us, I suppose.
| Thursday, 14 November 2002 | Echunga | Images for 14 November 2002 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Finally got round to fixing my Dell Inspiron 7500 laptop, which has been suffering from a broken hinge for five weeks now. The correct parts finally arrived at the Echunga Post Office on Tuesday, after I had to pay through the nose for them to be sent via courier. I am still anything but happy with Dell, but at least the repair went faster than I expected, despite their terrible documentation. Just remove the keyboard and the right-hand hinge cover slips right off and exposes the hinge. The whole thing took less than ten minutes. In the process, discovered that the hinges weren't made of plastic after all, just some soft cast metal:
|
|||||||||||||||||||||||
For reference, this hinge (the right-hand one, where the flat cable to the LCD display is located) is Dell part number 6307U, and the one on the other side is 7301U.
Also finished tidying up the office, which now looks much tidier:
|
|||||||||||||||||||||||
I wonder how long that will last.
| Friday, 15 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Up early this morning, and actually got some work done, even including installing SANE on sydney, the laptop I repaired yesterday. I bought a USB scanner years ago with the intention of getting it to work under FreeBSD, but never actually got round to it. Instead, Yana connected it to her laptop and booted Microsoft on it when she wanted to scan something.
Then to the airport to pick up Mark White, and to Gouger St. for a Yam Cha lunch. On the way out, witnessed a hulk of a man smash the rear window of a car with his bare hands. The driver did the right thing and drove away. It's difficult to see how the offender could get away: there were hundreds of witnesses.
Then picked up Liz Carroll, Gordon Hubbard and Stephen Rothwell at the airport. Dropped Stephen at a hotel, the same one that Jörg Micheel stayed at earlier this year, and with the others off home. Set Mark up with a connection to his home machine, and had a mini-meeting with Liz and Gordon, which kept us up until dinner time.
David Purdue came in round 20:30. Had dinner and a long talk. Late to bed.
| Saturday, 16 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Up earlier than I would have liked this morning, though I had only myself to blame: I had asked the Board to put the meeting forward by an hour. Into town and found iagu without difficulty. Andrew Rutherford had a wireless network in place, but we weren't able to get it working with WEP. The first key he gave us looked strange ("AUUG Board": it's not a valid key length). In the end we gave up and ran in the clear.
We had a pretty long agenda (thus the reason for the early start), but somehow we got finished with only minimal overrun. After that Dan Shearer came along and told us his ideas about running Linux.conf.au in Adelaide next year, with AUUG participation. It's not clear that Adelaide will fly; ACT is also interested, and we've taken the viewpoint that we'll let the LCA people decide the venue, and we're prepared to help at either place.
Off for a traditional drink after the meeting, but there weren't too many people left tonight, so left early. What a weekend. At least the hay is finally in the shed.
| Sunday, 17 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Spent most of the morning trying to catch up with the mountain of email which had accumulated over the last couple of days, and also started thinking about redoing the instant-workstation port to use Gnome instead of KDE. That was frustrating. Updating ports is really, really fraught with problems, and portupgrade doesn't really help much. We really need something better.
In the afternoon to a barbecue at Essey Deayton's place. She has now split up with Mark and has a new partner. It was pretty warm, but we had a good time.
| Monday, 18 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Back pretty much to normal today. For once, I'm roughly on top of my work, and spent some time writing up various AUUG-related activities.
Also started trying to use SANE, which I installed a few days ago. I'm beginning to think that there's a typo in the expansion of the acronym. It should be “Scanner Access Not Easy”. The documentation is terrible, it contradicts itself, and even the supplied programs can't agree as to whether the scanner (an Acer 620U) is detected or not. Did some ktracing and discovered that, although I told scanimage the name of the device to look at (/dev/uscanner0), it always tried /dev/uscanner (which doesn't exist) first. It opened and closed /dev/uscanner0 30 times, and finally decided that it wasn't there. Other tracing showed that it finally decided it was a completely different scanner before changing its mind and deciding there was no scanner there at all. The documentation couldn't make up its mind whether it needs microcode or not, and the configuration file (which didn't seem to get consulted) decided that it was a different model number. This will be a frustrating nut to crack.
| Tuesday, 19 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
As Yvonne was about to leave for shopping this morning, we discovered that the battery in the car had died. I'm really astounded how short-lived car batteries are in Australia: this one was 2½ years old. In Europe, battery failure is associated with cold, dark weather, short distances and congested traffic, none of which apply here, yet we're continually having to replace the things. Got Jim Couch to put in a better quality (well, more expensive) battery, which had the effect that the car radio wanted a PIN typed in. I've never known a PIN for the radio: last time they must have found it out for us. Yvonne ended up at the Holden dealers, where they wanted $15 to retrieve the PIN. Grrr.
Into town for the ADUUG lunch, where I spoke at some length with Michael Davies, who is interested in running the Linux.conf.au conference in Adelaide next year (2004). Maybe AUUG will be able to help.
After that, to Internode and finally got the machine that I was to set up as the new www.auug.org.au. Thence to Mount Barker and bought a filter for the sprinkler system, which has been consistently clogging up lately. Back home, tried to fire up the AUUG machine, but it seemed DOA. Spent some time considering how to approach the issue. I'm rather concerned about changing processors too frequently: putting the fan on them requires more force than I like, and I'm worried about damaging things.
| Wednesday, 20 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Guy Schubert here working on the sprinklers again today. That seems to take forever. At least we have the water filter now, which by the end of the day had already caught a surprising amount of grit. If the amount doesn't drop, we're likely to have to clean it once a week.
Writing about Vinum for my book, and tried to set up a 5-CURRENT machine on one of the disk arrays, with very little success. The host adaptors I have don't have boot PROMs, so I have to boot from another machine, and somehow that didn't work right. My current intention is to have one IDE disk with four BIOS partitions, one each for FreeBSD, NetBSD, OpenBSD and Linux. It seems, though, that I'm still being bitten by the 504 MB limit, which I thought we had left far behind us.
Also trying to set up the new AUUG machine, and found that the CPU appears to be defective. It's a mobile Celeron, and it has a core voltage of 1.6V. The Celerons in zaphod have a core voltage of 2.0V, and they (well, one of them) work fine in the machine. Looks like we'll have to find a new (well, different) 500 MHz Celeron, which could be a challenge.
| Thursday, 21 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More work on the machine upgrades today. Some of it went smoothly, but by no means all of it. Installing software is still not nearly as easy as it should be. Ran into multiple problems with the ports collection again: it seems that it assumes the content of root's PATH variable. Mine didn't have /usr/X11R6/bin, which caused imake not to build. Ended up completely updating all the ports I installed just yesterday. There must be a better way to do this.
Sent out a message to the local user groups looking for a 500 MHz Celeron processor. To my surprise, I got about half a dozen replies. Looks like that end of the problem is less than I thought.
| Friday, 22 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Got off to a bad start today: on IRC, setuid was grumbling about problems with the FreeBSD IRC stack and Palm Pilots, but he wasn't prepared to fire up gdb to investigate it, so I ended up debugging the machine, somewhere in Rhode Island, with the owner of the equipment, somewhere in Texas, looking on. Spent a couple of hours discovering at least that USB will happily remove an open device, and looked at how to fix that.
Apart from that, more work on the book, and tore apart the chapters on updating things again. This is all too painful, as I discovered while I was trying, yet again, to do a clean install of monorchid. Nevertheless, had the feeling that I was getting somewhere.
| Saturday, 23 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
On with the book today, writing about Vinum, in particular how to migrate file systems from disk partitions to Vinum. Things are not in very good shape. First, disklabel no longer allows overlapping partitions. In principle this is a good thing, but in the case of Vinum partitions it's a feature, not a bug. Fixed that, but couldn't commit it because we were in code freeze.
After that, discovered that I could create the partitions, but after stopping and restarting, Vinum rejected two of them. So I set up a debug kernel and discovered that my asf macro for gdb no longer works: it seems that the load address must now be a constant, whereas it used to be an expression. We can work around that, but after that I found that it only recognized the function names, not the local variables. Tried an older version of gdb, but it doesn't seem to understand the object format. What a day.
| Sunday, 24 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Spent quite a bit of time on kernel debugging today, including downloading the stock distribution version of gdb, which behaved in exactly the same way, except that it didn't understand the -k (kernel debug) option. That's not needed for serial debugging, however.
After some investigation, discovered that gdb does, in fact, load the symbols from klds with add-symbol-file: the problem is that the klds in /boot/kernel/ were at least partially stripped, but file still states:
/boot/kernel/vinum.ko: ELF 32-bit LSB shared object, Intel 80386, version 1 (FreeBSD), not stripped
In any case, as I realized later, reading the symbols from /boot/kernel/ doesn't make any sense, since that's on the debugging machine, which might not even be running FreeBSD.
The other issue remains: the new version of gdb doesn't accept expressions in a number of places where the old one did. My macro used to run through the linker_files list looking for a name starting with v, not exactly an elegant way to do it:
define asf
set $file = linker_files.tqh_first
set $found = 0
while ($found == 0)
if (*$file->filename == 'v')
set $found = 1
else
set $file = $file->link.tqe_next
end
end
shell /usr/bin/objdump --section-headers /modules/vinum.ko | grep ' .text' | awk '{print "add-symbol-file /modules/vinum.ko \$file->address+0x" $4}' > .asf
source .asf
end
The obvious alternative was to output the value of file->addresses to a shell script, but that seems impossible too. It's about time people scrapped gdb and wrote a new debugger.
Then it occurred to me that kldstat can also produce the information I'm looking for, so off to find a good script to produce the input file for gdb. I couldn't find anything which would do the job without significant effort, in particular handling hexadecimal numbers, so wrote a C program to do it, in the process reminding myself how little of that kind of programming I've done lately. It has the added advantage that it now loads symbol files for all KLDs, but I still need to find a way to locate them automatically: theoretically they're in the subdirectory modules of the kernel build directory, but in practice, if you use symlinks, it unwraps them, so my modules were in subdirectories of /src/FreeBSD/5-CURRENT-MONORCHID/src/sys/i386/compile/MONORCHID/modules/src/FreeBSD/5-CURRENT-MONORCHID/src/sys/modules/, 15 levels deep. I don't really see that that's necessary, but currently the program doesn't attempt to locate the files: you have to tell it if the directory is different.
Finally got down to looking at my bug. Turns out that vinum_scandisk doesn't check whether a partition is a Vinum partition or not, and in this case the swap partition overlapped the Vinum partition and was much smaller. That stuff needs a rewrite.
| Monday, 25 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
What weather! While the East Coast of Australia is in the grip of the worst drought in recorded history, the weather here seemed to have escaped from a Central European November: cold, foggy and wet. Echunga had 26mm rain, Meadows had 40. Since we're half way between, I'd average that at 33mm of rain, 5% of the annual average.
More work on the book and Vinum today. It turns out that recent changes in the operating system have caused previously benign problems to cause a panic. I really need to spend some time looking at the code, and after that we should have the additional benefit of a Vinum root file system. In the meantime, though, it caused a certain amount of pain.
| Tuesday, 26 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Still more rain, 41 mm so far this week. One effect is that our water tank is now full to overflowing.
Yvonne into town this morning, amongst other things to pick up a new processor for the AUUG machine. Grrr. It was another 1.5V part, and doesn't work in the motherboard that I have.
More work on Vinum, also uncovered more bugs. I'm beginning to wonder if something has changed in the way the system allocates locks: I create a plex, do nothing with it and delete it again, and its witness reference count has been decremented. We really need some hardware memory access breakpoints in the debugger. Finally got the Vinum chapter finished, and I'm relatively happy with it.
| Wednesday, 27 November 2002 | Echunga | Images for 27 November 2002 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
The end of the month is coming, so it's time to get my articles written for Daemon News. Spent some time on the ports trawl article, which is really more work than it should be.
Also did some debugging on Vinum and found the second bug I saw the other day. I'm still concerned there might be more. I suppose I should write a test suite.
| Thursday, 28 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Up earlier than planned this morning: Yana's car had a flat battery, and she needed help charging it. At least it meant that I had a lot of time, and I had finished my daily email by 9 am.
More work on the book. I should be working on my articles, but I can't get my head around them for some reason. At least I made some good progress on the book, and I have another chapter ready for review and a second following pretty soon after. Also more fun with the Ports Collection. I really should give up trying to write articles about it: I seem to have above-average difficulties just getting the damn things to compile.
| Friday, 29 November 2002 | Echunga | Images for 29 November 2002 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
No water in the toilets this morning. No, not another burst pump: since installing the water filter last week, we've been running the toilets off the dam water. Went to take a look at the filter element. It didn't look as clogged as last week, but it was still pretty impressive for a couple of days' running:
|
|||||||||||||||||||||||
The interesting thing is that most of that stuff appeared to alive. There are two kinds of
recognizable life forms: this one 
appears to be a kind of water flea, and this one
appears to be a kind of snail. Both
are tiny.
The reeds in the pond are also in good shape. I've stopped poisoning them and have resorted to pulling out the odd shoot that appears where we don't want it, but yesterday I noticed a whole army of little shoots popping up:
|
|||||||||||||||||||||||
The ones on the right are the reeds, and a lot are still under water. The plants on the left are supposed to be there, so it'll be difficult to poison them.
Phone call from somebody in Adelaide looking for FreeBSD 4.7 today. He apparently didn't know about the local BSD user group, and had found me via Google. Turns out that he lives in the same road as I do, about 10 houses down (which translates as about 2km). Small world.
More work on the book, more progress. I now have another chapter at review draft status, which only leaves about 4 to go. Also worked on re-porting X on sydney. I've continually had trouble with the freetype2 port, and today was no exception: it kept running (BSD) make instead of (GNU) gmake. Turned out it's hard-coded in the sources, and nobody has patched it. How could that ever work? After fixing that, though, I still wasn't out of the woods: gmake recursively started itself about 200 times before running out of processes. Grrr.
| Saturday, 30 November 2002 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More work on the articles today. Somehow I just don't have enough material any more for the Trawling the Ports Collection articles. I'm going to have to drop this one. The other article, Daemon's Advocate, was also a pain.
As if that wasn't enough, more and more work on the instant-workstation port. I'm still having trouble building, and put in Yet Another PR. Frustrating day.
Do you have a comment about something I have written? This is a diary, not a “blog”, and there is deliberately no provision for directly adding comments. It's also not a vehicle for third-party content. But I welcome feedback and try to reply to all messages I receive. See the diary overview for more details. If you do send me a message relating to something I have written, please indicate whether you'd prefer me not to mention your name. Otherwise I'll assume that it's OK to do so.
| Top of page | Previous month | Greg's home page | Today's diary entry | Next month | Greg's photos | Copyright information |
