Don’t wonder if I appear online less often these days and commit weird and plain ugly things. I’m young and need the money</asr>

Uh, EXPN? Sure. I’m currently working for a company and a customer who both originally were basically the local CCC meet cast into a company. I need to learn a lot of things, and my view of the world has been utterly turned upside down.

At least, I was able to spot some geocaches in the more obscure parts… some would call it primitif… of the world. And in the Elsaß.

Ah, X11… the input is passed as keycode, not as keysym… it’s not easy, trying to write a thing for entering unicode hex chars (damn US keyboard missing the “<>|” key). IME won’t go. Maybe fake a selection and a paste event. Compose works, but only in xterm, which grows from 8 MiB to 34 MiB size and 1 minute startup time, each, when expanding composure. I don’t think that’s worth it, even if unicode is big.

As a new “standard”, we’ll #define all functions we introduce to themselves, like arc4random_pushk(3), to rid autohell.

Still cats have less sorrows. Now it’s proven that feline specimen can actually spell. Poor “lolcats” who are abused by many.

Discovering Rivella and Tonnerre mentioning that swiss chocolate exported into the EU is artificially made more sweet than inland production (yuck, sweet chocolate) are some promising points voting for working in Switzerland. It’s still expensive and a major change though. But the payment is good, too.

While mkisofs is finally gone, I had to fix lots of bugs and port a new language, lang/pawn, actually better than others, but the interna still look more than very fragile. Same problem as with Ewe: written for Win32-only at first, made portable later by persons who don’t know how to code for Unix. Ugh, and… I’ve got to learn Python.

Either my VPN, gecko2s Server, or the network (at home, Strato, or here) sucks. Hm… maybe it’s the backup Tonnerre makes.

I should search for some more geocaces, but now that I have some more time than at the beginning of the week, the weather is worsening, and we still have the duties. *grml* I need tear set up, too…

I have just finished porting the aqua version of LyX (which I need for @work), which is going to become aqua/lyx-mac. Version 1.5.x needs Qt4, which we do not have on MirOS just yet, and the Mac version installs into an application bundle, so it is a completely separate port.

First of all, I discovered TWO bugs in Qt's build system. Not only are the generated .la files wrong, but the .pc files are broken, too. My first bug report contained lots of detail but what finally landed in their bad joke of a bug tracking system is a one-liner. That should teach me not to write bug reports longer than one line ... Seriously, WTF?

Next, our pkg-config (both actually, pkgconfig-bsd is not better) does not grok frameworks. In order to get a correct result, I had to port pkg-config-0.23. It is not in the trunk yet because I was told that newer versions "make bad things happen" on MirOS. What I will do is try to build some stuff using this newer version of pkgconfig and try to find out if it works. So, if you are feeling lucky: Install this and try to build some software on MirOS.

I started hacking on the bootloader issue, decided to move out the various libraries into src/kern/ first, and completed that for zlib, began it for the rest. The other places – src/lib/libc/ src/sys/lib/ src/sys/arch/*/stand/lib* – now look much cleaner already, and in fact can do some better. I also merged bootloader and other improvements from OpenBSD before beginning the file move dance, but it still worked for me in qemu for testing, so I didn’t expect much breakage.

Furthermore, MirOS now officially has OPTU-8 locale (no more CESU-8), even if masqueraded as “en_US.UTF-8” for better application support. The CSUR had already, as posted about quite a few times, allocated us a PUA range, which mksh was already using, and now the base system and libc are beginning to catch up. Thanks to a pen and paper “game” I have sort of a defined finite machine modelling the conversion states, and found out I can get by with the mbstate_t we already were using. tr(1) already doesn’t wreck ISO-8859-1 accents any more (nor binary files), although that definitively needs some testing.

I’d just like to warn people that HEAD (MirOS-current) is a pretty moving target at the moment.

… I still wish I were a cat, sometimes. Just have to stay clear of cat abusers.

Tonight, I managed to port Wine 1.0 from pkgsrc®, with thanks to Hubert Feyrer for mentioning, except it seems to stumble about W^X and memory allocation/mapping in general. I guess we need some help from OpenBSD toolchain experts and possibly both FreeBSD® and pkgsrc® porters dealing with this. I can’t, but I tried a lot.

My other project for tonight failed, once because it has just too many unported dependencies, second because one of them requires gcc4, namely webkit (non-Qt). Okay, they say webkit-Qt can be built with gcc 3.4.6, but it seems to be not recommended. Thus no midori for us ☹ (yet).

The third thing was getting rid of J�rg Schilling… but due to the heat I slept bad and could not get up early. Really! (Yeah yeah…)

You all should read Tonnerre’s wonderful blog posting about how he got rid of Subversion, luckily. (I wonder about said “new” developer’s preference as I’d rather use CVS for everything…)

… oder: de vitae felidae

Katze müßte man sein, oder Kater halt, je nachdem. Den ganzen Tag sich faul in der Sonne räkeln, unter dem Heck eines Autos liegen und ab und zu den an ihr vorbeiradelnden Zweibeinern zuzwinkern und sie mit dem Kopf verfolgen; wenn man Hunger hat, reicht es, den Dosenöffner auf zwei Beinen mitleiderregend anzuschauen und/oder zu -miauen… manche haben sogar ein eigenes Blog oder gar ein BSD nach einem benannt; wem das nicht reicht hält sich den Nachbar des eigenen Dosenöffners, nämlich homsn, als zusätzlichen Hausmensch (quasi wie wir uns Haustiere, z.B. Hunde, halten)… der krault einen dann, und man muß nichtmals hingehen.

I just coined a new term: IMACS. The connotation is: I’ll hack the ukbd(4) driver to allow inmetaaltctrlshifting entropy into the kernel too, soon.

Update 26.07.2008: sadly, USB keyboards don’t seem to generate any interrupt requests on modifier key presses or releases. This means you’re stuck to DIN or PS/2 pckbd(4) or Sun (&c.) zskbd(4/sparc) for entropy collection via IMACSing. Can’t be helped…

With the latest edition of MirOS #10-current’s snapshot (read the announcement for more details), we have trumped OpenBSD once more. While being dual-arch is nothing new (to them), and being runnable as Live-HDD, Live-CF-Card, etc. is nothing new (to us), we haven’t yet had a dual-arch Live-CD/DVD (or HDD or CF/SD/MMC/… Card or USB stick) yet. Well, only one of the two architectures has Live CD capability, but we neither have baselive image capability for sparc yet (though that would be possible, even if somewhat kludgy) nor would it have fit on the ISO 9660 image within the boundaries of a CD-R. And DVDs don’t match well with SPARCstation 20s ☻

Coming to that, I recently recherched if gcc 3.4.6, which we currently use as system compiler, can be used to build gcc 4.3 (with special care on the Ada part). Looks like you can even use gcc 2.95 for that, lucky us. This means I should somehow discover the Ada patch for gdb 6.3, which – again, luckily – we use, and apply it. I had talked with Benny about the compiler issue… originally I wanted to strip down the system compiler to the bare minimum once a more recent gcc (and possibly, LLVM, SUNWcc, pcc, etc.) are in ports and usable, but with the proposals to rewrite gcc in dreaded C++, C++ API issues, licencing issues, Ada bootstrapping, etc. we will probably continue to require GNU make for building the base compiler (the bad part) and be able to announce we’re good for developing (the good part of keeping the full-featured gcc near base). Just the Java™ part will not be re-enabled in near future, as it isn’t used for bootstrapping Iced Tea anyway. (It can be used for simple things though.)

With all the friendly fire, OpenBSD bashing, etc. aside, I would like to use this place to thank a few OpenBSD developers for helping me out with a bug in mksh as well as the issue with sshd(8) on sparc (and probably i386 even if it didn’t show up) I had (which even was my own fault). While you guys still don’t have an mksh port in your tree, some of you are really helpful (also hi Miod) sometimes, and I hope we can drink a few beer together at the next FOSDEM etc. and FrOSCon of course. (Which I hope we will also be able to use to discuss the state of our X11 tree.) To Linus: while OpenBSD developers may be masturbating monkeys, at least they’re able to design stable APIs, drink beer, and use sane version control systems with central quality control while being annoyed by humppa.

The usual unrelated news: Basel, ich komme! Schon mal GRÜẞE im Voraus an die Schweizer, die dies lesen.

mksh side notes: some older versions of SunC 5.9 fail on ({ stuff in acomp, use env HAVE_EXPSTMT=0 to build then. nwcc also has issues.

MirBSD side notes: We build for V8 CPUs on sparc, but still only V7 are emulated by tme. And while V9 CPUs in V8+ mode shouldn’t be a problem, there’s no interest from the OpenBSD camp to run the sparc (not sparc64) port on sun4u or even sun4v machines, so no fast build machines yet… *snief*

Since CacheWolf was the only use case, I decided to remove MirEwe from the repository entirely, as CacheWolf will be rewritten to use the Eve VM, inevitably.

Not only have I removed a last source tarball (distfile) of Lunix Ewe 1.49mb4, with CompileEwe.zip, JavaEwe.zip, and ewe.jar pregenerated, but the comma-v rcsfile(5)s are also archived in a gzipped sv4crc file, gzsig(1)ned, as usual.

And, also as usual, one more unrelated thing: config(8) -ef /bsd should not only be able to handle gzip(1)ped kernels, but also personalise them by rewriting the initial_entropy array. TODO. The snapshot, dualive ISO 9660 image, is almost complete and will be uploaded later.

Update 26.07.2008: personalising done, snapshot broken.

mksh – The iShell

Thanks to help from people in the various IRC channels, and gecko2@ jailbreaking his iPhone 3G (not G3…), I was able to run (albeit not test – there is no Perl for the iPhone yet) two mksh(1) binaries on it: a normal one and a fat binary (aka universal binary – this one, containing code for 5 (five) architectures, can truly be called an universal binary ☺).

I packaged the modified build script, an extra ldid binary (built for OSX86) provided by someone else, and the output of the build process, together with a readme, as a gzsig(1)ned tarball on our server again – note that the cryptographic signature is only for time stamping (and download integrity checking) and does not imply any warranties, especially for the third-party software involved.

While here, we would be pleased if the reader supports our proposal and requests Apple to include mksh in the default installations. We would also be pleased if the mksh-iPhone binary were to make it into one of the Cydia repositories, contact us via IRC or mailing lists for assistance.

If someone were to package Perl 5.8 or so, that’d be great too.

To the readers from ##iphone-hackers on the one and #iphone-dev on the other IRC network: *waves a friendly hello*

My plans are as follows: after squishing out all remaining bugs, I’ll publish another MirOS-current snapshot (i386 baselive, i386 installer, sparc installer, some selected few binary packages for both arches), announce that on the website, and make a CVS repository tarball.

After that, I’ll concentrate on setting up the new servers (tear and euynome) fully, and dispose of rant-featuring-herc. Features I wanted, such as full AES vnd(4) encryption, DJBDNS with no split v4/v6 transport binaries, etc. will have to wait.

With the infrastructure in place, we can start hacking again. We had planned on bringing out MirOS #10bis in time for FOSDEM, but it looks more like just another -stable snapshot, and maybe another month for the second release of -stable; Benny will have to work some on the branch prior to its tagging (I’ll concentrate on -current), especially because we want to use X11 (and gcc) from HEAD, since these are not branched.

mksh – The iShell

Try this:

	$ cvs -d anoncvs@anoncvs.mirbsd.org:/cvs co -PA mksh
	$ cd mksh
	$ cat >iBuild.sh <<-'EOF'
		DEV=/Developer/Platforms/iPhoneOS.platform/Developer
		SDK=${DEV}/SDKs/iPhoneOS2.0.sdk
		CC="${DEV}/usr/bin/gcc-4.0 -arch arm"
		CPPFLAGS="-I${SDK}/usr/include"
		CPPFLAGS="$CPPFLAGS -I${DEV}/usr/lib/gcc/arm-apple-darwin9/4.0.1/include"
		CPPFLAGS="$CPPFLAGS -F${SDK}/System/library/Frameworks"
		CPPFLAGS="$CPPFLAGS -F${SDK}/System/library/PrivateFrameworks"
		CPPFLAGS="$CPPFLAGS -F/System/library/Frameworks"
		LDFLAGS="-L${SDK}/usr/lib -L${DEV}/usr/lib/gcc/arm-apple-darwin9/4.0.1/lib"

		export CC CPPFLAGS LDFLAGS
		exec "$@"
	EOF
	$ mksh iBuild.sh mksh Build.sh -r
	$ file mksh
 

Mach-O executable arm

Of course, you need the iPhone 2.0 SDK on your Mac for this (and, presumably, an installed native mksh… otherwise just change the above instructions a little.

We haven't tested it yet, but if gecko2@ can figure out how to run mobileterminal.app inside the emulator, he will… and possibly jailbreak his iPhone G3.

Maybe this fact (availability of mksh) strengthens our point.

In the meanwhile, gecko2@ has found the geocaches I placed.

Update 18.07.2008 12:42 – we have a fat binary with five architectures (amd64 arm i386 powerpc ppc64) now. This is the iShell, indeed.

Okay, so I installed Windows® 2000 and Interix again. And UWIN. And the Microsoft, Digital Mars, and Borland command-line compilers. And mksh worked, of course.

Then I tried DJGPP. Yeah, mksh(1) for MS-DOS®. *sigh* GNU bash, the only shell available, SIGSEGVs on me trying to build anything nōn-trivial, even worse with gcc 4.2.3 than with gcc 3.2.3… but even pdksh could not be made, due to SIGCHLD missing. Too bad, no mksh for you, Mr “I don’t accept bug report mails from your IP, nor in the webinterface” Delorie.

The iPhone… will have to wait. We can install the BSD Subsystem and the Terminal application in gecko2’s SDK’s ARM emulator, then build and test for that. I just gave him the link to the BSD package, but he went to bed shortly before, so maybe tomorrow.

And then, mksh R35b can go out, we fixed some bugs and tested on more platforms (fixing testsuite shortcomings during that time, even).

As requested, Benny has written to Apple, Inc. about inclusion of mksh. Lucas “laffer1” Holt of MidnightBSD has said to follow, as has Andreas “gecko2” Gockel, the Fink maintainer for mksh. Link to this entry.

Lucas actually said he would care more about and use mksh(1) if it were integrated in Mac OSX, as most MidnightBSD developers (like Benny) do most of their work on a Macintosh. He has integrated mksh in his operating system quite some time ago.

mksh is also the default shell, including /bin/sh, on FreeWRT Embedded GNU/Linux.

Andreas already has run tests for using mksh(1) as /bin/sh on Darwin, although I did not request that personally, and brought up the question of /etc/profile adjustments and ~/.mkshrc integration. (FreeWRT has a changed system-wide profile including a system-wide copy of the mkshrc file; other possible ways would involve setting $ENV or (preferred) placing .mkshrc in the New User’s Skeleton and in root’s home directory, if existent.

Strong points of mksh: free (as in BSD, now even without advertising clause), small, fast, portable, easy to use, can run many ksh93 and bash scripts, is actively developed, and benefits from the BSD development style (central development, security focus).

To everyone with an Apple ID: please suggest to Apple to include mksh(1), write to them how good it is despite lacking popularity, how compatible it is, and just how much you like to use it. Even if you do not use it yourself, if you can follow our reasons, write to them. If you have access to other operating systems, do the same, especially with mere porting frameworks, as the OpenBSD ports tree still does not include our port, which was ready-made for committing by me to reduce workload for them.

Some unrelated side notes… Geocacheing continues: – Dr. Pfeffer especially liked mine^WWaldemar’s Zaurus SL-C3200 with CacheWolf running on Ewe… even if I still think it’s dead slow. Interestingly, porting Ewe to the iPhone G3 would not violate the clause preventing you from creating “instant messaging or real time navigation software”, and CacheWolf itself would not have to be ported. I’d like to have more RAM on the Z though – maybe via the SD Card slot?
SCSI sucks. And mksh gets better… IRIX, and a bug fix. Just still no SunOS 5.5 (missing /usr on the HDD).

This is the title of the bug report I filed with the Apple Bug Report Tool yesterday. The problem ID is 6069931; I think you can only see your own bug reports in the tracker though, so you may not be able to see my report. FWIW, I filed it against Mac OS X 10.5.4 (the current version), with "Feature (new)" as a severity. For reference, the full text of the bug report follows:

11-Jul-2008 11:58 AM Benny Siegert:

I would like to kindly request the inclusion of mksh as an alternative shell into future versions of Mac OS X.

mksh (the MirOS Korn Shell) is a Korn Shell derivative under a BSD-like licence. It includes the features from the OpenBSD ksh as well as additional bash compatibility. Almost all bash scripts run under mksh, however with a speedup of about a factor 2. The code was checked for security problems by Coverity, Inc. UTF-8 support is fully integrated. With all these features, it is still very small, just 240 KiB as an i386 binary. The build process is done by a script called Build.sh, Mac OS is already a fully supported and maintained platform.

This could potentially replace the ksh93 included now as /bin/ksh or be installed side by side as /bin/mksh. In short, it can be regarded a more modern replacement for older ksh releases.

The homepage of mksh is located at http://mirbsd.de/mksh. The current version is available from http://www.mirbsd.org/MirOS/dist/mir/mksh/mksh-R35.cpio.gz.

Since quite a while I’ve had anger about BSDstats – their ports reporting stopped including ours (MirPorts Framework, NetBSD® pkgsrc®, OpenBSD ports tree, Debian dpkg), the maintainer was rarely reachable, at most by eMail, and the entire system had issues. Now, the site no longer exists, apparently nobody knows why, and our systems error out. Well, I have disabled BSDstats, this will probably make MirOS appeal more to privacy-conscious geeks again, and save us from getting more errors in our logfiles, especially the Live CD reporting was already totally broken due to weird User-Agent tricks.

I’ll now count on people just using IRC or so to say hi to us if they’re using MirOS, are happy or tell us things we could improve. Or meet us at conferences and pay for our beer ☺ and other beverages (we’re not BeerFSD after all).

Configuring name-based apache vhosts with SSL and IPv6… is hard. Really. This is probably due to it being “a patchy webserver”. Here is a quick cut and paste solution, well, sort of.

Scenario: I have a web server running name-based vhosts on the SSL port. I get a either wildcard certificate, or – in my case – a certificate with some subjectAltName extensions set; CAcert supports these now. All vhosts I wish to serve must be included, and the primary commonName (as part of the distinguishedName) ought to be the host’s primary FQDN, and it must be duplicated as subjectAltName.

I wish to serve a default page (just the standard index.htm from /var/www/htdocs/) to everyone who does not provide a proper host name (e.g. speaking HTTP/1.0 without Host: header, or just probing my IP addresses), and a couple of vhosts to everyone else. I use one IPv4 and one IPv6 address. The vhost configuration is placed in several files in /var/www/conf/vhosts/ called «vhostname».«port»; if some information is to be shared (e.g. vhost on both port 80 and 443), these are placed in «vhostname».common and included by the port configs.

Important: Do not use wildcard vhosts, they won’t work and give errors. Do not use the _default_ vhost in any form, it silently won’t work.

You might want to place directory information in the vhost configuration file instead of in the main configuration file.

Part of httpd.conf is:

	ServerAdmin «mail address for the server itself»
	ServerName «main FQDN»

	NameVirtualHost 192.0.2.1 443
	NameVirtualHost 2001:db8:1234:1::1 443

	<IfDefine SSL>

	<VirtualHost 192.0.2.1:443>
	Include /var/www/conf/vhosts/eurynome.443
	</VirtualHost>

	<VirtualHost [2001:db8:1234:1::1]:443>
	Include /var/www/conf/vhosts/eurynome.443
	</VirtualHost>

	<VirtualHost 192.0.2.1:443>
	Include /var/www/conf/vhosts/call.443
	</VirtualHost>

	<VirtualHost [2001:db8:1234:1::1]:443>
	Include /var/www/conf/vhosts/call.443
	</VirtualHost>

	<VirtualHost 192.0.2.1:443>
	Include /var/www/conf/vhosts/default.443
	</VirtualHost>

	<VirtualHost [2001:db8:1234:1::1]:443>
	Include /var/www/conf/vhosts/default.443
	</VirtualHost>

	</IfDefine>
 

(using documentation addresses)

Make sure every vhost except the default one sets ServerName (and ServerAdmin if it differs), and DocumentRoot etc. as usual. The default vhost does not need to set ServerName, but the line
ServerAlias *
must be present and it must be the last vhost in the list, as above.

If you still have questions – especially now that even OpenBSD has seen the light and applied the IPv6 patch I not only applied to MirBSD some four years ago but also suggested to them some two or three times – come to IRC or drop us an eMail, and we’ll try to fix things. Apache is definitively historic ballast, but still common enough on todays Unices.

So the uppercase eszett – ‘ẞ’ (or ‘ẞ’ if your monospace font has it) – is now an ISO 10646 standard. Your favourite BSD has been one of the first to add support for it to libc, a (contributed) keyboard layout, and even Markus Kuhn’s fixed-misc Unicode fonts, after the proposal with a real codepoint assignment came out. Not all Germans, nor typographers, but historicians, might like it, the discussion has been flamewar-like. But we knew this all along.

Ah, by the way. No, we aren’t dead. Websites are irrelevant, but if you end up here, please read about mksh anyway. Real questions are asked on the mailing lists or in IRC. And we don’t have a handbook (yet), because with two developers, nobody has yet had enough spare time to write one, plus you’d need a couple of handbooks – one for people coming from Open/NetBSD (Lite-based), one for people coming from Free/DragonFly/MidnightBSD (386BSD-based), one for people coming from Unix®, one for the Apple followers, one for the GNU/Leenocksers, etc. You see?

And while here: Xcode 3.1 (on gecko2@’s Macbook) comes with llvm-gcc… but not in /usr/bin/ but in /Developer/usr/bin/. It compiles quickly, produces fast and small code… kind of nice. If Apple brings LLVM+clang into a usable shape to replace our gcc 3.4.6 (gcc4 has unsafe optimisations that cannot be turned off, and all other compilers produce bigger code, which breaks the installation media (especially floppies) and SPARC kernels), in a form usable as compact command-line compilers (and not just libraries for Xcode integration), I’ll be happy. (They’d just need to add mksh to Mac OSX then to make me even happier, but I guess I’m sort of late with that request. Even if it’s fully free – in contrast to GNU bash – and, in its most recent incarnations, even advertising clause free.)

While “the swiss are laggards, and the people from Bern are the slow motion of switzerland”, other things take their time as well. However, vnd(4) updates have started, and other encryption schemes (including a hardware-accelerated AES variant for Columbo0815 from #IceWM) will end up in the code very soon, as promised.

During hacking, I spot other things – bugs in the random(4) code, as well as improvements to the kernel and XFree86® possibly leading to my pcengines.ch ALIX.1c board to be supported (thanks to undeadly.org for keeping me informed). And Benny fixed ports bugs again, as well as put security updates for Freetype2 et al. into the tree. Thanks!

FWIW: Opera 9.50 segfaults on me. It can’t be updated thusly.

This year's LinuxTag will not go by without the BSD projects exhibiting their fine software. Of course, we will attend – the usual gang (Benny and me, gecko2 and Przemek will help out). New is that we'll also be distributing a couple of MidnightBSD CDs and flyers I had a helping hand in creating, as they are US based and won't be able to make it to Germany for this event. This all would not be possible without the help of allbsd.de, again. Thanks, Daniel!

The MirBSD CDs distributed contain a Live + Install CD, as usual, of an unnamed #10-stable snapshot, plus a couple of packages, including Benny's new firesomething port instead of opera-linux. The MidnightBSD CDs aren't as big and miss the install part, but are a relatively new and especially stable 0.2-CURRENT snapshot (including mksh of course).

We'll attend Tuesday (probably late) till Saturday.

Last time I had tried qemu+kqemu (on GNU/Linux), it made BSD segfault, kinda like VirtualBox nowadays. But things seem to have changed: I ran the Linuxtag 2008 Live CD Edition inside it (on MirBSD) with no problems at all. Nice, and shows that BSDs can have LKMs.

Now missing: vmxnet source code… or evil BLOB? I suppose the latter.

Unrelated side note: Netcologne is cool, but they suck: you can only change the PTR for your static IP, which you pay 4.90 € per month for, if you’re a business customer. No IPv6 anyway (but SixXS has a new PoP, out of all places in Düsseldorf *evil grin*). And Strato gives gecko2@ two IPs, but only allows the MAC of the built-in NIC (so no bridging to domU or VMs). Why can’t they ever do something not totally clueless? (Okay, I could ask the support… Netcologne support even phoned me back after my eMail enquiry apparently went through their entire house until they could figure out what I need (to mail Fefe, who blocks PTRs like foo-www-xxx-yyy-zzz.ISP), but rejected; Strato support is, from my past experience, hopeless.)

The MirPorts Framework, which also runs on MidnightBSD, is the first to bring native firesomething support to said operating system, almost effortless thanks to our good portability infrastructure, and because bsiegert@ has already done the dirty work porting it to MirBSD. Thanks, Benny! And thanks Lucas for the chance to prove the superiourity of our infrastructure. And ctriv, don’t take it too hard, just take our patches and put them on your mport.

Hardware sucks. The Z doesn’t power on after it has been shut down for a while. The CF wi(4) is broken. Software sucks. And we all know it. A web forum sucks. They probably will never learn it. Yet people support their… attitude. Americans suck, but that’s not news. Luckily, there are a few not totally clueless ones. Life sucks, and bureaucracy sucks even worse. And there’s no way around that. Too bad. (I had an encounter with bureaucrats again… could you tell? Hah. They want money back, even though it belongs to me.)

Update 21.05.2008: Benny has committed an update to firesomething-2.0.0.13-0, and I bumped the dash ver and made it work as well on MidnightBSD. Sample build:

 $ cd /usr/mirports/www/firesomething
 $ mmake install clean PKG_CONFIG_PATH=$(mmake show=PKG_CONFIG_PATH):/usr/local/libdata/pkgconfig

You must have installed fontconfig from mports (and, of course, perl, for MirPorts to work at all.

This beast now works on MirBSD/i386 and MidnightBSD/i386, should work on OpenBSD/i386, OpenBSD/amd64, OpenBSD/alpha, OpenBSD/powerpc, OpenBSD/sparc, OpenBSD/sparc64, and possibly MidnightBSD/amd64 and MirBSD/sparc (not yet on MidnightBSD/sparc64, but they’ll want to test and fix that). No idea about Darwin and Interix. Benny is going to bump to the latest upstream version now, whereas Jonathan recommended us to go to 3.0b2 for increased speed and decreased resource use. We’ll see. The mports maintainers can now take our patches and get rid of their linuxulated firesomething binary port ☺

We have lived for 25 years with the seekdir bug, and even if it’s now on slashdot this does not mean we will immediately patch it. Besides, bringing that part of libc in sync with OpenBSD will involve libc and libpthread shlib version major bumps, which is a bit overkill for this diff.

It will go anyway with the upcoming merge of more recent OpenBSD base code. There are more pressing issues. But I have looked into it.

Happy birthday, laffer1!

Whew. We have a new qemu port, but it doesn't boot MirBSD/i386 any more. It almost boots MirBSD/sparc now though. Luckily I could backport the new port's ability to use kqemu. Thanks to Fabrice Bellard and the OpenBSD ports guys for this.

The Zaurus... SL-C3200... it's a neat device, but after I managed to get the pl2303 LKM cross-built (bah!), along with the usual tools mksh(1) and jupp as well as pax(1)mirabilis for the ability to extract my CacheWolf profile onto a (FAT) filesystem - did I mention just how much busybox sucks? - Ewe doesn't work. Meh. Go to OpenBSD/zaurus, enter a SIGBUS in gtk+1.2 - exactly what I want. The gtk+2 version works, but now, X.org plus GTK+2 plus Ewe (Java™ Ranz!) eat up all of the 64 MiB RAM plus initially 4 to 8, later 20 to 40, MiB swap. Not nice. (Some people are said to use a CF card for swap due to it being faster than the internal HDD, a microdrive. Geez.) If I'm cycling with more than say 5 km/h, the moving map hinders parsing the GPS symbols... NMEA 0183 at 4800 bps. But I managed to find another cache before going totally crazy.

After Linux (grml, FrOScon edition) has destroyed my NTFS partition, now the ext2fs driver has managed to impale quite some of my FFS filesystems again. This time, Linux isn't even at fault. (Hm, fsck_ffs(8) and e2fsck seem to have a common ancestor.) The downside however is that my encrypted home is now gone. Totally gone, as in, fsck deleted the dirent, inode, and allocation of blocks.
Yeah, I have backups, but only for the most important things, not for everything, and quite old ones. I have had to restructure my storage use anyway, now's a good time for it. Hah! And that after I chose to use ext2fs ipv msdos for a common data xfer partition between BSD and Win2k due to msdosfs eating up long filenames when a directory has a lot of them, even on OpenBSD 4.3/zaurus FWIW.

While I was quite reluctant to hack anyway recently, or even to idle in IRC or, worse, Jabber, I guess this'll throw me back even more. MirBSD isn't affected, except in further development becoming delayed.

There will most probably be a snapshot of MirBSD-current really soon. But other work (mksh R34, mirmake in Debian, MirEwe, ports, gcc) is postponed; I will try to focus on the more important things (AES vnd(4), improvements for CAcert, merging OpenBSD-current maybe even, but that not before the migration to tear is complete). OTOH, for both MirPorts on !MirBSD and pkgsrc® on MirBSD, I have laid foundation for others to build upon. Maybe they will.

Just a side question: why don't things just work every once in a while? And why, oh Murphy, do things go wrong the worst way possible, always?

On an unrelated note, mksh needs people running the current development version, to prevent mishaps like the one with the fullwidth characters causing wrong text output. To do that, use AnonCVS like this:
% env CVS_RSH=ssh cvs -qz3 -d anoncvs@anoncvs.mirbsd.org:/cvs co -PA mksh
Then, as usual, cd mksh && (sh Build.sh && ./test.sh -v) 2>&1 | tee log.txt to build and test it. Then, especially if it fails, send the logs to me.

Ewe is an embedded VM for some Java™ 1.1 compatible stuff. I had to fork it to use it on MirBSD and to be able to fix it. So well. It now builds on gecko2@’s Macbook. Using MirMake, of course *g* It’s even usable… which means that he can now use CacheWolf and his laptop to go geocaching.

Speaking of geocaching: more stats bragging… although a little different this time:



Although only the first two are CacheWolf committers…

Ah, damnit. It freezes the usbserial (Prolific) driver when accessing the serial port. Well, Apple… they don’t even use GNU as(1) either.

… as you could get it. This is because gecko2@ asked for some more cvs commit eMail “spam” ☺

While on the spam topic: do not send an eMail to one of the following addresses: junk@mirbsd.org, junk@66h.42h.de, or junk@mirbsd.de, or, again, <junk@mirbsd.org>, <junk@66h.42h.de>, or <junk@mirbsd.de> (greytrapping)

Luckily, my internet uplink has been stable for more than 4⅓ days now, after repeated phoning (an 0800 number then) and resetting the NTBBA.

Some more statistics:

Geocacheing continues: – now I’ve hidden my first two traditional micros (easy series), and one of them even is sort of a “lost+found” directory virtual cache overload.

BOINC continues, I’m in 9 projects now:
While only 8 projects show up at the moment, this’ll improve once the last project delivers in a result (it were more projects actually, but some don’t even work on hephaistos…) – now the first WCG valid WU returned from MirBSD! (MidnightBSD can’t, because I can’t run brandelf on the signed binaries of the apps… sucks to be FreeBSD derived ☻☺)

Ah, and, by the way: XTaran did not like external links, especially not secure links (https) in my wlog entries, so Planet Symlink doesn’t get them now (as it pulls via RSS), but you can look at it on the wlog.

… but calling does. The bad part is that the internet support number starts with 0900…, which means it cost me 1.69 €/min (and the call took me about 3‥4 minutes), but on the other hand, they quickly fixed the issue, sent me an SMS that they did it (just, sadly, not what it was), and since then, I haven’t been forcefully disconnected any more. This is good.

Now I’ve just got to look if this is stable, then re-measure my bandwidth. And in May, I’ll get even better upload (736 kbps instead of 608 kbps).

Verpeilungsfaktor stories: in our weblog source, we use RFC822 style header lines: “Date”, “Author”, and… “Title”. Not “Subject”. And I wondered why the posting headline wasn’t shown… Since we’re on Planet Symlink now, I try hard to find a matching one for each posting, because it looks stupid there with just the date.

Another one: we use four dashes on a line by themselves for separating log entries; the mksh regression test uses only three, as you can see. And yes, I fell for it (gave me a nice Perl error message over there, and a nice shell script induced error over here).

A last question (rhetorical: no answer needed, just nudging your brain): what do you do with your spare time, if everyone who’d to something together with you either moved away, is working or ill? And: if you were to move away yourself, would the situation improve, worsen or stay the same?

The ttftot42 utility I asked bsiegert@ to port was actually mentioned in some GNU groff documentation as a possible way to use OpenType fonts in ps/pdf manpages. I always liked the Gentium font (which we use for the MirOS CI/CD), and starting from when I first saw it, I hated anything remotely looking like the Courier font, so I guess this means that the PDF version of mksh’s online manpage will soon have a new look.

On an unrelated note:

	Apr 21 20:09:06 herc ppp[28555]: tun0: LCP: deflink: RecvTerminateReq(206) state = Opened
	Apr 21 20:09:06 herc ppp[28555]: tun0: LCP: deflink: LayerDown
	Apr 21 20:09:06 herc ppp[28555]: tun0: LCP: deflink: SendTerminateAck(206) state = Opened
	Apr 21 20:09:06 herc ppp[28555]: tun0: LCP: deflink: State change Opened --> Stopping
	Apr 21 20:09:06 herc ppp[28555]: tun0: CCP: deflink: State change Stopped --> Closed
	Apr 21 20:09:06 herc ppp[28555]: tun0: CCP: deflink: State change Closed --> Initial
	Apr 21 20:09:06 herc ppp[28555]: tun0: Phase: deflink: open -> lcp
 

(not using pppoe(4), to enable debugging)

For me, this very much looks like my ISP doesn’t like my nose or something and wants the ADSL (ppp(8)+pppoe(8)) session to terminate.

Another unrelated side note: MirOS ports/math/boinc definitively is able to return valid results to some projects (although some of these which work require setting the primary platform to Lunox, but I’m trying to get that bug on their server-side fixed.

Just returning from a week in the south, and feeling well. By sheer verpeil0ring, I was away the whole time that tg@ spent in Basel, just a few kilometers from here. But those holidays were necessary, and I enjoyed them very much. @work, there are new stagiaires of which I don't even know the name but it does not look as if anything had moved during my absence ;).

I am trying to port ttftot42, a nifty utility for converting TrueType fonts to PostScript Type 42 (which seems to be straightforward) and, more interesting for me, can create AFM files for them. Using those fonts in Teχ is only an afm2tfm away then. The last version is from 1999 and needs freetype 1. The author—very responsive indeed!—says this is not yet fixed but promised to send me a hg snapshot. While here, we now have a mercurial (hg) port ;).

Today’s pcc from anoncvs bootstraps successfully and builds mksh just fine, and is amazingly fast (almost en par with Microsoft’s compiler). Wow, they fixed all the things I ranted about in my earlier postings. Congratulations, pcc team.

Now tcc and TenDRA/Ten15 (schizo) are next (mainstream compilers failing). And LLVM/clang and ACK deserve testing.

I wish ranting would help with my internet connection… gotta fight with the ISP/Telco now.

I already ranted about pcc… well, I got a reply to my first mail to the pcc list (where the second one cleared up the five things mentioned in the previous posting), a sort of still friendly one-liner, to which I replied with that he should probably read my other mail, to which I only got an unfriendly comment that “you are wrong”. Hah! (Well, I got my “pcc -E” fixed.)

I guess I just cannot recommend to use pcc, and will have to maintain my own set of patches. Trying to get them upstream shipwrecks on a barrier of incompetence, regarding not only autotools but also how a compiler (cc(1) standard interface) must work: at first, on -O (or -O*), pcc did simply an Oflag++; which I mentioned as wrong (adding a fix)… but look for yourself. Oh, and they reply using weird – OpenBSD (latin1) or Windows (cp1252) – encodings on mails properly sent using Unicode (UTF-8), as is the default in sane operating systems like MirOS and Plan 9. Incompetence whereever you look. This matches the interesting UCB hack in mv(1) I recently found… or OpenBSD’s inability to port GNU tools.

Also, I suppose my ISP/Telco is going to get some angered tg@ tomorrow. The NTP Pool scores show that I’m suffering from a lot of network hiccups. This LCP fluctuation kind of sucks, as does the current transfer rate. Just the latency is still surplus, 11.2ms to heise.de (suckers as well, but for totally different reasons) and 18.3ms to google.com (also suckers, for a couple of yet another reasons I think I already elaborated).

I probably could compile mksh on pcc again… if pcc would compile itself. Hey, this one is about the contrary of OpenBSD or lynx, where development versions are stable… pcc should warn before cvs upping.

Every time I try to build mksh with pcc, either it’s totally b0rked or I have to fix it. Today: ragge doesn’t understand autotools. (He added a test to configure.ac which ① gives a syntax error when failing, ② fails when compiling pcc using pcc as the compiler, ③ doesn’t show up if /lib/cpp exists (on GNU/Linux, I suppose), ④ produces a broken cpp(1) executable if it fails, ⑤ doesn’t even test for the thing it is supposed to check.)

Oh, and the charsets of the mails are b0rken. (WTF windows-1252 when I send UTF-8?) See all the ugly details here (XXX insert link).

Back home… there just ain’t such place as [::1]… (that’s localhost for all of you who don’t use BSD). Swinging on the bike and going to the ice dealer, the best of them all. It was kind of nice in Switzerland and it’ll be a hard decision for me whether I’m going to move there or not. But after arriving at home, past the bike tour, I fell into sleep pretty soon. Travelling may be interesting, but it sure is tiring.

Too bad I couldn’t find the two geocaches I looked for on the way back.

I hacked some mksh on the train, until I had no power left in the batteries… the laptop literally just went off all suddenly… and continued that until now. We have some quite interesting new features in now, only sad point is that we still can’t hexdump NUL.

I should definitively get my new server (tear) running now, for which the only dependency left ought to be the updated vnd(4) crypto stuff. This will take a while, as I’ll design a new on-disc format as well for improved security (think of keys, IVs, and so on). After that’s done I’ll give y’all a snapshot of MirOS-current, and update a lot of ports.

Maybe I should work on bringing a regular sparc boot floppy into the tree as well – last time, I had to hand-craft one. But it will be lacking.

There’s so much interesting stuff to do. Working on the Zaurus, ALIX, my SPARcstations (still no big monitor yet, so I couldn’t test Miod’s patch to make tvtwo(4/sparc) work yet), more FreeWRT devices… but I can’t neglect my dayjobs either. And I ought to learn to read and fix Perl *sigh*

This sucks: I have network (internet) outages since last night. Sometimes, ppp(8) + pppoe(8) still work when pppoe(4) doesn’t, but most of these times, both are unusable. The rest of the time, I sometimes have huge lags. My ISP (which unfortunately is a telco, but they aren’t completely clueless either) wanted to upgrade me from a 4 Mbit/s connection to a 6 Mbit/s connection as the old product doesn’t even exist any more (and I’ll save 10 €/month now), and the cable length (230m) isn’t an issue either. Testing today (as per the salesman I should have it May 1st, per the acknowledgement mail April 15th) I’ve got about 6 Mbit/s down, but my upstream speed is even reduced! WTF?

Argh! Later on this night, my network connection is so flakey…

This time I did find a geocache far away from home…

… in contrast to when I was in Bruxelles, as Benny and gecko2 didn’t seem to want to have time for that (or walk at all, they coerced me into the tram). This time, I went cacheing with Tonnerre, and he kind of lined it.

Time to push opencaching in Switzerland. He said he might even drop some caches (although – jokingly I hope/suppose – his first idea was „Finding Sandro“, where the cache is a person… or his home appartement). Likewise, I’ll push OC (and, a little, TC) whereever I’m going to live or lived.

While here, special greetings to the TGIF@BS meeting which I won’t attend, as I’ll take an earlier train back home tomorrow. It was nice here, much more so than in, for instance, Berlin.

Perl is evil. But knowing the basics of other programming languages helps. I guess I’ll invest some time into learning perl better, so that I can get rid of it (in MirPorts, for example), and better understand what others try to write in it (so it can be converted to mksh if possible, or at least fixed or optimised).

People can be quite annoying at times (mostly in Jabber, but also via eMail or IRC). Hey, if I just don’t reply my current location per eMail, sending another one asking specifically for it again isn’t going to improve my mood. Neither is constantly annoying me with enquiries about whether I’m really gonna move („zügeln“) here or not, after I had already stated I’ll think over it next weekend (or so), since I have a few reasons pro et contra, some of which are orthogonal to what I see here. I concentrate on getting a feel right now. Oh, and texting me one messager after another in Jabber (or, worse, by SMS to my Natel) even if I don’t reply (which, on the other hand, does not imply I’m willing to conversate either!) just gets on my nerves. And: go fucking RTFM, and don’t fucking bother me with „the XXXXU2B controller doesn’t exist, because the vendor website only lists the XXXXU2W“ – if you know any vendor websites you should long know better than to trust them.

For what it’s worth: for building MirEwe, you need very current MirMake (at least 20080411), g++, GTK+1.2, GTK+2, libjpeg, zlib, and their development headers. It should work on GNU/Linux and the BSDs for now. No platform other than i386 has been tested yet, but I’ll take on the Zaurus running OpenBSD, I guess, as I finally got the uplcom(4) working. Ah, and to rebuild the class libraries you need ecj and paxmirabilis/MirCpio – I did the ecj part on Debian and the rest on MirBSD.

XTaran mentioned that there is a Secunia advisory for mksh: SA29803. Wow. This would then be my first one. But people really cannot read: the advisory states it appies to “Secunia Product #18328: MirBSD Korn Shell 3.x” – WTF? I mean, version numbering surely is difficult these days, with Linux and related software often having four or more decimal points(!) in them… but I reported it to them using their web site form.

I don’t know why Secunia made this advisory, but I suppose the Fedora person has his hands in there, as he told me yesterday that he needs an advisory for marking the update as security fix in their package system. I told him that this is not a lack of wanting to document, or even lack of communication (skills), but that I merely don’t know how this kind of issues is usually handled “out there”. I haven’t dealt with GNU/Linux any more since I started using OpenBSD back then in 2.9 times, and then, this procedere wasn’t widely used.

Ah, speaking of OpenBSD. They are sometimes not even at fault when a bug report is mis-communicated, even if some people don’t believe it. And they don’t do the major/minor game either. They just still have the decimal dot for hysteric raisins.

Actual user feedback matters. My fork (MirEwe, just to annoy Jonathan) of the Ewe VM now has a fix, or rather workaround: when opening serial devices, “/dev/” was prepended to the device name. Now if users enter “/dev/ttyUSB0” (leenocks) things don’t work – but this is hard to debug. MirEwe now warns and does the right thing. The warning is in there for portability ☺

Ah, and: I could finally log my first (locationless, though) TerraCaches. That makes me an active user of all known platforms. Even if it cost me an entire night of re-reading a book I knew from a looong time ago. But actual caches will follow, and I’m even already planning to plant some myself.

Basel is getting nearer. I’m still not too happy with the option to move away from here. We’ll see – at least I’m going to take a big chance when I see one, and peek into how it is over there. Even if Benny is on Malle.

How can a bug like this be unnoticed for so long, while the two source code lines in question were specifically touched and diffs reviewed by big names such as deraadt?

Of course it’s me who has to fix longstanding bugs from 386BSD, VIA C3 AES data corruption bugs in the OpenBSD kernel, LP64 bugs in OpenSSL, etc.

Da wir ja alle eine große Familie sind, möchte ich hiermit Grüße (Grüsse?) an die Schweizer Leser, die dieses Weblog (nein, ist kein Blog) auf Planet Symlink entdecken, senden.

Ad auditorēs qui nōn possunt legere linguam germanicam: Salvete amici Cōnfœderationis Helveticæ!

Ich bin übrigens Mitte der Woche in Basel zu Besuch, Benny ist leider zur Zeit auf Mallorca, Urlaub machen. (Hoffentlich fernab vom Kampfgebiet, äh, Ballermann und so.)

As you can read, quite a few new versions of our portable software has been released. Well, sort-of-portable, but for nroff I plan on improving, and MirMake 2 will be a lot better too, kinda like mksh.

Geocacheing continues:
Even Jonathan got hooked now. We found another (his first) two today.

As we’re sort of a big family, you’ll occasionally find German-language postings here. Don’t wonder. We have people who especially read that.

Ah, and while we’re at stats bragging:

These were made on gecko2’s Intel Mac (Darwin), my laptop (MirBSD), and a few even on hephaistos (GNU/Linux). MidnightBSD needs to use brandelf, thus execution of Linux/i386-ELF binaries fails or, after branding, checksum verification (which BOINC does) fails. Sucks to be FreeBSD derived.

I’ll be in Switzerland (Cōnfœderatio Helvetica) next week, while Benny happens to be on vacation on Mallorca (hopefully not near the war zone called Ballermann). Development may slow down a little due to that (no more 100+ CVS spa^H^H^Hmails per day, yay!) but will not stall.

I haven’t written here for a while, but I just want to get one thing said. I might have done what Sun feared first thing after I got a Java™ VM (even if it’s only single-threaded Ewe for embedded systems) running stably on MirOS… it now has a native method arc4random_pushb(3)… yeah!

Entropy is determining a lot of my life these days anyway. For example, accidentally sleeping too few, too much or at weird times, being phoned by random people, talking with Vutral about further possible improvements in our RNGs, being asked by CAcert.org if we are suitable OS for their high-security boxen (almost, but we’ll fix the missing parts, and some they’ve got to do themselves as no off-the-shelf OS does), and reporting in huge masses of entropy to CAcert.at Research Lab – some 128 MiB samples with Firesomething “Bon Echo”, as Opera ISE’d out and Lynx just first ate up all CPU then none at all any more without monitorable activity…

Ah, and BOINC is running stable on MirBSD. MidnightBSD has some issues, mainly due to the fucked-up FreeBSD kernel and brandelf.

When Mozilla Firefox(TM) aka www/firesomething version 1.5.0.10 came out over a year ago, I tried porting it (of course). However, not only did the build not finish, it filled the entire filesystem before failing. At the time, I was updating from something like 1.5.0.4 so I did not know which version exactly caused the problem. Thus, I made incremental updates up to 1.5.0.8, where I seemingly lost interest. Now, with a faster build machine, I continued the updates, discovering that 1.5.0.10 is really the version that fails.

A diff between firefox-1.5.0.9-source.tar.bz2 and firefox-1.5.0.10-source.tar.bz2 is 10 MiB. Most of that is in CVS directories, which are included in their fucking releases, believe it or not. If you leave out those changes, the diff is still 5 MiB and 155000 lines, of which 90% are in security/. Mind you, the whole source tarball, from calling the directory mozilla instead of adding a version number to including CVS directories to using several different build systems, leaving unused configure scripts in the tree, etc., positively REEKS of a flagrant disrespect for those that build from source. All this seems to be meant to encourage you to use official Mozilla builds—which is fine if you happen to be on one of the few supported platforms: Windows, Mac OS X, and Linux i386. We are not.

Anyway: in their infinite wisdom, the Firefox developers chose to upgrade the included NSS libraries (Netscape Security) from 3.10.x to 3.11.4 in a minor security update. This version sports extensive internal restructurations—another nice way of saying "fuck you, porters". Thank you, Mozilla project. Of course, the NSS update is not explicitly mentioned in the release notes.

This tech note says: "The low-level freebl cryptographic code has been separated from softoken on all platforms. Even on platforms for which there is only one implementation of freebl, there is now a separate freebl shared library. The freebl library implements a private interface internal to NSS." This new library is the core of the problem. After the NSS libraries are built, they are cryptographically signed by a program called shlibsign which, in turn, dumps core, generating a 3 GiB core dump in my case!

I had suspected a problem related to our security features, especially W^X. The page about Building on Fedora Core 5 says: "For those with SELinux in enforcing mode, you are likely to run into problems both with the shlibsign during the build process and with the running the final build related to SELinux denying execmod permission ..." However, with some difficulty, I managed to build a debug version of everything for analyis with gdb. During the start of shlibsign, one of the init procedures loads the native freebl (?) module, which promptly loads itself. On and on, until memory exhaustion.

This comment in security/nss/lib/freebl/Makefile provided a clue to the solution:

# The blapi functions are defined not only in the freebl shared
# libraries but also in the shared libraries linked with loader.c
# (libsoftokn3.so and libssl3.so).  We need to use GNU ld's
# -Bsymbolic option or the equivalent option for other linkers
# to bind the blapi function references in FREEBLVector vector
# (ldvector.c) to the blapi functions defined in the freebl
# shared libraries.
ifeq (,$(filter-out BSD_OS FreeBSD Linux NetBSD, $(OS_TARGET)))
    MKSHLIB += -Wl,-Bsymbolic
endif

Adding MirBSD to the OS list fixes the build. And it works!

Now, firefox 1.5 is old and unsupported. So why was this update important? It makes further updates possible. The same bug was holding up my long-finished firesomething-2.0 port. I already have a package for firefox 2.0.0.13 but the port needs just a little more work. I am also confident I will be able to provide a working port for firesomething 3 when released.

Wow, my first posting in this new weblog. My laptop, a new MacBook Pro, spent about 10 days over easter in order to really do a MirPorts bulk build. 1503 binary packages for MirOS #10 have been uploaded to the mirrors. This has also been an occasion to review some largely untouched parts of the tree. Most of the problems seen are related to missing distfiles, changed download URLs, etc. Some packages did not build during th bulk build but when run non-recursively, they worked.

Some bugs in the resulting packages remain: You must manually install the expat package for most of the stuff to work. gtk+ insists on writing to /var/db, which makes it unsuitable for AS_USER builds. firesomething only when works as firefox. Still, this gives you a wide variety of pre-built packages for the new release. Enjoy!

I seem to have a sort of fan club, mostly related to mksh, but also to jupp. Interesting.

Benny has built 1503 binary packages, amounting to about 954 MiB of data. Uploading…

We just got an email about Metalinker, which looks interesting enough to try, from its primal developer Anthony Bryan. We now have it ☺

This night, my internet connection failed every so few minutes (returning a PAP failure, despite me being in midst of a session). I took the chance to upgrade herc’s software (kernel, userland, a few packages) and spotted that sendmail(8) bug. I also took the chance to rebuild the two RAID 1s (one for /, the other for /MirOS and the CVS repositories), which took me a while and about a dozen attempts, but finally I succeeded.

Other than that, I didn’t hack much. I walked about 5‥6 km (single way) to buy myself a Döner though ☺ bad weather doesn’t have to mean stay inside.

We should be at way more than 1500 downloads in total by now… 1405 on the Germany 1 mirror, more than 100 on Germany 2, an unknown lot on Japan, more than 60 BT downloads from me and friends… that accumulates quite.

Happy Spring Equinōx to everyone!

I merged the crypto improvements to HEAD: we now have improved Rijndael CBC code for UVM swapencrypt, and this code uses the VIA C3/C7 PadLock™ ACE if existing – whose code contained a now-fixed data corruption bug.

Next on the TODO list is: make vnd(4)’s encryption use the same code if AES is chosen as cipher algorithm. Allow selection of Blowfish (stay compatible), AES-128, AES-192, AES-256; other algorithms may follow. We will need a new keyfile format and stay backwards-compatible, but this is not a problem.

We got a PUA assignment from U+F900‥F97F (tentatively) for our encoding proposal, and should use “one of the various non-characters” for the NUL encoding. But: “Emoticon U+FDD0 is actually Unicode for the eye of the basilisk…” – U+F000‥F7FF are now reserved for Linux’ straight-to-font map (and some subranges are used by Windows® and Mac), and F800‥F89F for Mac (and possibly, Linux). – Although I got scolded again for chosing a 16 bit wide character type, I believe this compromise with all its good and bad sides is the right way to go for us at the moment. As for how to codify this, I still do not have a final answer, but I think using wrappers for SUSv3 functions which cannot fully support the proposal is a good idea. Most use cases should work with the SUSv3 functions, anyway.

I still haven’t ported mksh to BSD/OS 3.1, OSF/1 V2.0 and Ultrix V4.5…

I’m impressed – 1377 full downloads from the Germany 1 mirror, add to that the partial downloads (wget -c or so-called “download managers”), these from the Germany 2 and Japan mirrors, BitTorrent (which is 36 times from myself, and a couple of dozen times from my friends, and even more from the unknown peers) and you won’t think of us as irrelevant any more.

I’ve hacked on the crypto improvements branch again. I decided that getting tear up and running would be my priority target. It turned out to be a great way to hone my programming skillz as well: I got a lession about pointer aliasing.

Except the actual VIA C7 part, I now tested it quite well on i386-qemu and sparc, where the latter took about one third the time compiling… the new CPU and RAM pay off.

Oh yeah, and hacking on stuff always points out unrelated bugs… such as NO_GZIP=Yes for kernels not working, or <bsd.lkm.mk> being out of date (I had to MFC that, even). And that src/kern/z needs tender care – the transition is still not finished even for zlib.

I hope the discussion about our charset/encoding proposal will find a solution… I feel like constructing a bikeshed, as you get about that much feedback there too ☺ Benny said he doesn’t grok the SUSv3 functions, like mbrtowc(3), enough to follow – and I can totally understand that. Maybe a Unicode guru person like Markus Kuhn can help – I asked him.

My primary SPARC build box demo now has a HyperSPARC 150 MHz ROSS CPU, instead of a SuperSPARC 75 MHz, and 512 MiB RAM instead of a mere 128 MiB (and with that, twice the RAM of my primary i386 box).

But then, I got three SCSI HDDs from the same stone age, and each of them has a different connector. This sucks.

gecko2@ just called me. He operates www.mirbsd.org and got a little surprise on his daily traffic report. Summing it up, his server and myself, and a friend of mine, together, adding HTTP and BitTorrent transfers, have had about 500 GiB traffic in the 3½ days the #10 release is now available. Alone the HTTP direct downloads of clients that got the ISO in one piece (not HTTP 206) number sits at 861 at the time of this writing (850 five minutes ago, 825 ¼h ago).

As an immediate measure protecting his server against being taken offline for traffic limit trespassing, I redirected to allbsd.org for direct downloads on getting.htm and suggested him to install bandwidth throttling/limiting for apache. Don’t be surprised.

Update: half an hour later, it’s at 867, so I think the change of the direct link helped. Sorry for the inconvenience at both gecko2 and our downloaders – but then, to the latter group: You should’ve used BitTorrent anyway.

Oops. Changed the wrong link. 923 downloads… 927…

In addition to my primary geocaching site, OpenCaching.de, and the commercial crap site, GeoCaching.com, which has most users, I now also registered at TerraCaching.com, which is a semi-closed site providing “high-quality” caches. Stats bar gallery:

Of course, I didn’t find any TCxxxxx caches yet. I need to get fully registered (“sponsored”) at the site first. This isn’t a big issue tho.

Update: I’m in. Nearest caches are in Blankenheim, Neuwied and België…

MirOS ξ (MirOS xi) is not only our eleventh release (as I started counting at zero), it is also the Unicode release. While it finally makes sparc a fully supported platform, the real focus is on Unicode, and bsiegert@’s girlfriend seems to have realised that better than I did.

However, in fact, that was only a start. We need to change the character set in order to be able to handle binary files transparently with Unicode-enabled applications – col(1) and tr(1) in MirOS #10 – before converting more applications to use Unicode.

Please give feedback on the thread linked above, if you can.

MirOS #10 does come with everything needed by applications for full Unicode support though, and in contrast to OpenBSD, things really do work. This justifies calling the release like this, even with a Unicode character in its codename.

For the first time, the “tag line” comes without a “WTF?”. You may take this as a sign that we are not confused about ourselves, have gotten over the initial cause to make MirOS and now no longer merely are a team that wants to improve OpenBSD. We are a small but powerful operating system project, with goals (already met or new ones) of our own. We still track OpenBSD, but that’s no longer the focus. Benny’s girlfriend also got that right. And her mouse did attract the users, and stayed topic on the other BSDs’ websites. Wow.

The MirPorts Framework can now be used on MidnightBSD out of the box, no special patching needed. MirMake, MirCpio, MirCksum and the package tools have been updated appropriately.

However, MirLibtool support is still missing. This is trivial, I just don’t have the time to do it right now, as I should be doing the release at the moment. Still, feel free to play with it.

Yeah. I did port MirLibtool to mnbsd, and found a bug (in upstream, even) while doing it. Nicey. Both devel/glib1 (autoconf-2.13) and devel/libtool (autoconf-2.61) seem to work (regarding libtool build systems, shared library building/naming, etc.) for now. I’m a little bit unsure about shared library naming conventions, but you can follow FreeBSD’s, even though Todd Vierling and I think they are weird and/or not what one wants. (For insiders like Benny: they do not use current/age/revision, set the third number to 0.)

We have a new “p5” port module, which must now be referenced by any users of ${P5ARCH} which do not include the “perl” module.

pvalchev@obsd replied to my issues with the UMAC64 hash that I might want to report the bug in libgcc’s umoddi3 implementation upstream. This shall be my TODO, someone please remember me to do it (check how to reproduce on $common_os, with $latest_gcc, etc).

miod@openbsd is helping me with the tvtwo(4) card – unluckily, I’m unable to test diffs because my monitor is too small… Doesn’t anyone happen to want to give away a TFT that can do 1152x900 for free? ☺

As written in my earlier entry (still to be ok’d by benz) about GPLv3, editors/nano is okay (for now). This means that rsync 3 is probably ok to go in now too (especially since we have no patches). It’s said to be much faster and less RAM-hungry, which is especially nice.

Besides from my TODO on MirPorts (and the portable subprojects) and MidnightBSD, and the release engineering process, and my other want-to-do hobbies like hercules(4) wsfb(4) support (and an XF86 module, and emulation support for a HGC in qemu), and a couple of other things, Waldemar has made a point:

MirOS definitively needs to shift away from “we want to make OpenBSD better, and we do X, Y and Z” towards “we want to do X, Y and Z, specialise on W, support V, and while doing all that, we are of course as secure as OpenBSD and track their goodies, and by the way, we have GNOME and Frozen-Bubble”. This would give the MirOS Project an actual face, which could attract users and development capability/potential. (And it would imply a re-design of the Flyers’ and website’s content…) We’ll have to think about it, but he is probably right.

The new cksum port sets a variable HAS_CKSUM, which will be used really soon now in the MirPorts infrastructure to replace all the old cruft (_CKSUM_A, _CKSUM_SIZE, _HASHES). From RSN on, you will either have our current cksum(1) from MirOS #10 or MirPorts, or you won’t, in which case it will only use the “cksum” algorithm of the OS’ own cksum programme (which is rather ubiquitous).

We have a new sample file, “portmdoc”, and I’ve converted yet another manpage to be fixed with regards to GNU groff, like I did with mksh(1) after the R33 release. Expect this to continue.

The FSF is now mistaking that lazy moronic finnish student’s excuse for a patch management system for a version control system as well… the config.{guess,sub} files are now in git. Yuck!

Our kernel now should handle signals wrt the extended i386/amd64 ABI fine: thanks to the Debian GNU/kFreeBSD developer Aurelien “aurel32” Jarno, the direction flag (DF) is now cleared on traps.