Sometimes it’s good to double-check. Benny has found a bug in my fix to gnu.port.mk, leading to incorrect dependencies. I couldn’t get the sparc floppy to boot — several NetBSD® users had the same problem, leading to a posting about how to clean a sparc floppy drive, luckily. Note that I am lazy, that’s why I just tested the floppy in my SS10 instead. This gave me quite a hard shock, though.

I prepared the Mini-ISO, and — new — a Midi-ISO containing the floppy images as well (if extra netboot kernels were required, they would be also on it, but boot.net is enough), which will take the place of cdrom10.iso on the download area, while the Mini-ISO will sit on the release ISO.

This means we’re now all set to release today. *crosses fingers*

Yay! I got a SPARCstation 20, a SPARCstation 10, and a SPARCstation 5, several CPUs and sbus cards, an external HDD, an external CD, and some more toys, from wbx@ — he has no sparc parts left now. Sad, as he is the initiator of MirOS/sparc. Good, because I can cluster them now. I could test the framebuffer, XFree86®, even build binary packages with distcc if Benny ports it (I still have to get on friendly terms with distcc first though, and I won’t use it for official releases).

Being the better OpenBSD sometimes sucks. For instance, if our make(1) implicit rules generate grammar.h from grammar.y, and not y.tab.h, as a certain lex.l wanted to include. Well, that could be fixed — re-rolled ports10.ngz with new games/bsdgames.

I’m building a number of selected binary packages on the sparc and very few basic ones for i386, to add to the release ISO. (More binary packages can be downloaded from our mirrors later.) This is holding us a little, but I’ve got plenty of time over the weekend to carefully finish the release engineering process. Better than having bugs, eh?

I actually found a bug in gnu.port.mk during porting to mnbsd; gecko2@ had found it too but I couldn’t reproduce it back then.

Sorry for all the delay in the release announcement and publication, but first we have to get the ISOs ready — since they contain i386, sparc, source and CVS it’s a tad difficult and quite time-consuming.

Ha! Giving support in #cvs in IRC and being called life saviour, master, etc. is really giving me the chills, in a good way.

I’m exhausted, will continue hacking on the release tomorrow.

I’m trying to hand-bake a floppy for the sparc though. It won’t contain an installer, but enough to download an installer onto a HDD. Ports have aborted (gnupg can’t import keys, but decrypt them just fine — weird), but I’ve got enough for tomorrow.

It’s really annoying. This is actually a pre-release issue. There is a port, editors/nano, which snuck in before the release, containing GPLv3 licenced stuff. This means that I either have to remove nano from the mirports tree for #10 or read the GPLv3 and allow it into the tree.

Looking for OpenBSD’s opinion on GPLv3 in ports, I found that my “dear friend”, the oksh developer, has struck again: his oksh-0.3 is GPLv3’d now. (At least he did not remove the UCB copyright from some files.) And to add insult to injury, the description of oksh at Softpedia, which is almost certainly provided by my “dear friend” Henry Jensen <hjensen@delilinux.de>, pretends that mksh(1) is not Bourne compatible (which it is, at least no less than oksh) and buggy, which it definitively is not. He said that, at the time of his forking, mksh was buggy, but could not tell me one single actual point. His only complaint against mksh was that it does not support the GNU bash-like $PS1 hackery. *sigh* (Did I mention that oksh-0.1 violated the MirOS licence on mksh?)

Phew, I read through it. And I think we have a sort of official stance on the GPLv3 (even though bsiegert@ of course has to agree to make it truly official). I still don’t like it, but it could be worse, and for ports we do not need to care that much.

Today, I helped wbx@ to clean up his basement: DECstations, SPARCstations, VAXstations, Alphas, mac68k, hppa, SGI O2 and Indy, and some other boxen either went to the recycling company today or will tomorrow. gecko2@ retains some mac68k and maybe one hppa, I try to get some 32-bit sparc parts, and bogus got an E450 (heavy!) and an Ultra1 for toying around with, as he now exclusively runs Solaris.

It’s annoying that not every CPU works in every SPARCstation — for example, the two SPARCstation 10 we found have 33 MHz CPUs and 400 MiB HDDs, one has OpenBSD on it, one Solaris 5. Nice. Tomorrow we’ll look at the two SPARCstation 5 which are still there. The monitor sadly doesn’t work. Maybe one VGA conversion cable from the SGIs will help, but I don’t have a monitor capable of doing the sunfb 1152x900 (maybe with a zx(4) card…)

The SPARCstation 20 of mine (“demo”) has finished building the release kernels and tarballs, so I can proceed with the release engineering process ASAP. Like I said, we’ll do a unified medium. As with i386, the preliminary dist sets may show up on the download mirrors already for a while…

Building the release took about 6h22′ at 1 GHz on i386, and 81h22′ on 75 MHz on sparc, for what it’s worth.

Für lange Zugfahrten o.ä, wo man keine Lust hat zu hacken, ist ganz gut dies hier geeignet: z.Zt. 128 je ca. 200 Lynx-Seiten umfassende Harry Potter Fanfics, die ein recht logisches, aber manchmal nerviges Universum, das jedoch sich streng nach dem Original richtet, aufspannen. Einfach runterladen und dann offline lesen.

Ich bin ja ehrlich gesagt immer noch am sicken, daß Freenode mir einfach den Nick „lynx“ geklaut hat… jetzt kann ich gar nicht mehr so deutlich anzeigen, daß ich am Schmökern bin. Merke: wer „linked nicks“ hat, sollte diese trotzdem alle 29 Tage mal touchieren, da man im Staff nicht nachfragt, selbst wenn der Hauptkontakt gerade online ist.

I’m still pissed about Freenode taking the nick “lynx” away from right under my arse, having been online with the main nick of “linked nicks” (the NickServ feature), just because I had not used that nickname for a while. Note: touch all alternative nicks every 29 days. RichiH says it’s policy to at least ask the owner, but seemingly they don’t.

Oh, another wlog entry ☺

I found out that, actually, you can commit into both a branch and HEAD within the same directory on the same commit command. It’s just our log scripts which suck. This will be fixed RSN. It works if you use another, different directory as the last one (like I thought). This also means you can ignore commitid 10047D8245653D8184A and 10047D8248447FC1F5F, as these never went live.

Someone wants to do an appliance with FreeWRT, maybe I can let him hire (and pay) me for that. I could definitively need the money at the moment (hint hint).

If not, I can always port MirPorts to MidnightBSD…

… and I actually did, this will lead to some changes, even a few real improvements not only for mnbsd but for mbsd too.

This is probably the last entry in this wlog. If you are using the RSS feeds, you might either want to subscribe to http://www.mirbsd.org/wlog-10.rss, or change your subscription to the new (as of yesterday) feed at http://www.mirbsd.org/wlog.rss, which is a symbolic link to the current developers’ weblog.

A fix for the ppp(8) security vulnerability has been committed to MirOS #10-stable already. This is still in time for the release, and will be included in the source10.ngz dist set as well as in both i386 and sparc base10.ngz dist sets. Thanks to TNF’s replaced, who mentioned the issue to me in the #!/bin/mksh channel in IRC.

The i386 part of the #10semel release is already built, but, oh WTF we don’t have a codename for this release yet. Well, the sparc part is still compiling, which leaves us another few days to think of one. I’m already uploading part of the sets, so that the distribution process will be quicker later. (Some of the files will change later on, though.)

Benny already has written a release announcement draft. Like I said, he is my best man. He didn’t even attempt to list the changes, which might even be the better approach… In the meantime, I’m updating some of the web pages already — you can, in theory, netinstall #10 now, but some links will have to be fixed later.

I still urgently need a SCSI host adapter for my new server tear, U160 or U320, SCA LVD, 32-bit PCI, no RAID, supported by MirBSD (i.e. OpenBSD 3.5), as the current one, donated by wbx@, has issues. Or a fix. Plans to migrate to tear have been postponed due to this problem, which makes the progress in the project seriously suffer.

The code is almost frozen, we’re only waiting for a few updates that should still go in. Benny allowed me to add the ‘u’ to the flavour variables of MirPorts, after a year or so of thinking about it ☺

I fought the Leopard today, but I won: we can now use our own getopt(3) suite on Mac OSX 10.5 again, and some other bugs and minor issues have been fixed too. Plus, all the hashes, including ADLER32, SFV, SUMA, TIGER and WHIRLPOOL, now work on Darwin too (tested on both i386 and macppc).

Now it’s release engineering. I have started tagging, after a few final updates and fixes, and asked Benny to write a release announcement. Let’s see when the builds are done, calculate a good four days for the sparc side. Afterwards, we’ll switch to wlog-10, and back up the repo to get it all on one CD (i386, sparc, source, CVS).

Thanks Benny, I’m glad to have a developer like you too.

Whew. mksh is done. I hacked a little on mcabber, and if your libc is recent enough, it even can display the new uppercase eszett the DIN proposed onto us (ẞ ← it’s in fixed-misc in MirBSD, and in a few other fonts such as the Linux Libertine). Mine wasn’t, of course. My work system is, actually, too old… (currently I’m building i386 stuff either in a chroot or in a dual-boot).

Now that tg@ is back, MirOS has certainly regained its momentum. Great work, it is good to have a developer like him.

For the last six months, we always distributed the same snapshot on exhibitions. I am almost ashamed to admit that the CDs we had at FOSDEM 2008 were burned from the FrOSCon 2007 ISO image that I had lying around on the showcase box, schaaf.mirbsd.org. So my radical proposal was: the code we have has had a lot of testing already and has shown no show-stopper bugs (except on the new Intel Macs, but that is another story). So: let's release what we have and call it MirOS #10.

However, we urgently needed to apply some security fixes first. My task was X. I managed to update the included freetype to 2.3.5, which contains many security-relevant bugfixes and is recommended for all users. Alas, many places under xc/ have to be touched for it to integrate well. I also applied the X.org patch from OpenBSD 4.2-stable. I had to do the applying by hand as there seemed to be some coding style changes from XFree86 to X.org but the code itself seems to have been mostly unchanged since 1987 or so. Some of it does not even have ANSI prototypes (in XFree86 at least).

Some ports have also been updated, mostly security fixes. My build machine is now MirOS in Parallels on a MacBook Pro. I have aqua/qt4-mac in my tree but it does not install quite right yet. A new, unified Ghostscript port is also in the pipeline. However, convincing the build system to only use the system zlib proves difficult.

After pushing out the mksh distfiles, ports, etc. (and, of course, spotting a couple of bugs too much for my taste), I finally found time to work some more on my TODO. rm(1) now can do random overwrites including file (and directory) renaming to random values before unlink(2)ing (and rmdir(2)ing). Thanks to TNF, again, for some of their code and bugs.

Another thing is putting the installer onto the Baselive CD. This would not easily be possible as of now, I see, because it uses, for example, some hard-coded paths like /mnt and commands like umount -a -t nonfs, which could cause normal operation on the rest of the running Live CD environment to cease. So this will not be in the tree in time for #10semel. Sorry, Benny, it was a good idea we had.

I also upgraded OpenSSH, but un-did the unifdef(1) -DBSD_AUTH change, as we need a non-BSD-auth version of sshd(8) for bsd.rd. During the process, of course, bugs were spotted… e.g. in the docs. Of course, the new internal sftp subsystem isn’t on the ramdisc, either.

And, I upgraded Sendmail, without the support of OpenBSD…

Note: new kernels and old /sbin/sysctl and /usr/bin/uname, and vice versa, will not have the kern.ospatchlevel entry, which is required to build XFree86® and/or imake(1) with MirOS.

Why do I always find bugs at almost after release time? Why do I let my coffee become cold, have too few milk at home, etc? Why do I sometimes forget to mount -uw /x before qemu’ing an image placed on that filesystem? Why do I sometimes forget to plug in the power of the laptop and then wonder why it suddenly turns off, having no APM nor ACPI?

I’d like to call a vote for MirPorts: replace the FLAVOR and FLAVORS variables by these with correct spelling (FLAVOUR and FLAVOURS), keep it as it is, support both with English preferred, support both with American dialect preferred? I’m strongly tending to British English and would like to have this general policy in MirOS.

OpenBSD/zaurus works, although it’s a PITA to get a working Linux 2.4 on it first. The USB NIC (aue(4)) doesn’t work, though (not enough power); but it at least tries, under OpenBSD, while it is ignored under Linux 2.4 and 2.6; my USB stick works on the Zaurus but not on the laptop or the WL-500g; a CF card reader only works on the Asus box, so I must use that for data transfer (and cannot use ntpd(8), which sucks) with smallish 16 MiB and 32 MiB CF cards from my digital camera…

And why do some bugs found past the last minute need so lengthy attempts at fixing because it cannot be done inside the current implementation?

I managed to ssh into all possible boxen to test whether the upcoming release of mksh will work on them (well, the Debian GNU/kFreeBSD qemu instance had to be upgraded first). I even installed some Linux 2.4 and then OpenBSD on wbx’ Zaurus since nobody of them could give me a shell.

Still some testing to do, but I’m okay with what we have, not only for mksh(1) but also for the entire OS. Expect #10 finally next week or so.

OpenBSD is weird. They did, in 4.3-beta locked-tree state, link OpenCVS to the build, but no fixes for the necessary sendmail upgrade or IP ID vulnerability yet (or even planned).

Schon mal geröstete und schokierte SOJABOHNEN gesehen? Hmm… die schmecken ja sogar. Ziemlich herzhaft.

Now I’m fixing Debian fixes to upstream software. Funny. But people who re-invent strdup(3) do belong shot, even if their crapware is public domain. Luckily, mgcc(1) catches sizeof (char *).

I’m done with ${foo/bar/baz} support in mksh(1); the mcabber port can now do umlauts (stupid implementation, they should use more native functions, especially since they do link against libncursesw).

The MirPorts main page is now connected again. Rob “rjek” Kendrick is currently porting his small, fast, CSS capable, JS planned, NetSurf browser to MirOS (many thanks here!).

Benny is doing good progress on the XFree86® security fixes.

Dang, it’s much too late again, but I hacked some cool things into mksh, which were planned for some time; now I need someone to test this. Also, here strings (hd <<<$str) are now documented (and do NOT work like in GNU bash unless $str is double-quoted), whereas the replacement substitution — ${foo/b+(a)@(r|R)?(\/*)/baz} — isn’t. This should, however, further improve people coming from other shells.

Benny is still working on security updates — now: X11. I should better get my ports and ssh etc. upgraded too. If I just knew what was important now… we need actual user feedback for this, I think.

I wish OpenBSD’d update their sendmail(8), so we can do that too ☻

We actually distributed about 182 CDs yesterday, and today, early morning, no CDs are left any more. We’re also about to run out of flyers. Benny’s showcase box aided really in catching users (especially the pink mouse, for the female part of the visitors, who tried Frozen-Bubble when we offered it to them).

Benny and I also decided on Unicode improvements and hacked Mac OSX Leopard support into MirPorts, this time correctly. This is really weird, and both of my fellow Mac fanboys decided that Leopard is quite buggy and thusly sucks. Probably due to the UNIX2003 certification?

Bruxelles is fun, and the hotel was very good (not really cheap, but not too expensive either). Sadly we probably don’t have any time to really go geocacheing, and our usual restaurant — le petite planéte — no longer exists, although we found a suitable replacement, with a delicious Assiette Mergues. Belgian beer was as good as usual, although it makes getting up early on the next day more difficult. The OpenBSD people donated 3 cans of Jupiler to us, which was unexpected but makes room for hopes of improvement.

The frietjes booth outside actually understood my attempts to talk to them in dutch and, although my pronunciacion is very bad (of course), she told me it’s not too bad overall (for a duitslander). They seem to pronounce ‘ie’ as a long ‘i’… I need more practice, but this part of Bruxelles is mostly french only.

Today, I merged back the cvs repository from hephaistos and tweaked both the ChangeLog and idcache as well as the ,v files to have sort of correct Commit IDs. I'll have to go through a few files worth of diff, but that'll follow next. Benny has done good work, and also thanks a lot to gecko2.

The process group IDs have been preserved in the lower 32 bit of the new commit ids, the upper 44 bit are, as usual, the time.

We'll resume normal-heavy operation very soon, so keep your INBOXen ready :)

As a side note, sorry for the inconvenience. As you might have already read, “herc” had a hardware failure (broken keyboard controller chip, due to a screw short-circuiting it), and it was “resurrected” inside of “rant”, which cannot, however, be easily booted (as the discs were IDE, and it can only boot from SCSI or — via the fxp(4) card — PXE). Due to my away time, I was unable to restart the box after it hung up. I was, as of now, unable to finally transfer services to “tear”, the new box (VIA C7), because I still need a working SCSI U320 (or U160) controller. The planned-impending (but then delayed) transfer of services to “tear” caused me to not fix the “herc” hardware first (as it would not be worth the effort).

We will be working on restoring all services onto “herc”/“rant” within the next days, try to get a SCSI HA ASAP, and build an improved (and possibly more failsafe) infrastructure.

The unplanned away time causes a late #10 release, and it will not be in time for FOSDEM, but it will come. Also, some things (nroff, Unicode) have been thought through once again, with interesting results, which will lead to further improvements in these areas in the future.

As an update to Benny's first wlog entry mentioning a possible mksh bug I have to say that this was fixed in mksh R31b: The “unset” builtin always returns zero, even if the variable was already unset, as per SUSv3 (reported by Arkadiusz Miskiewicz via pld-linux → oksh)

The 20080218 release of mirmake is the first in about four months. It contains one important bugfix:

• Mac OS 10.5 ("Leopard") and Intel Macintosh machines are now supported.

This new Darwin version is now UNIX2003(TM)-certified; to achieve compatibility, "symbol variants" are used for some functions, such as getopt(3). This, however, breaks the getopt implementation that comes with mirmake. Thus, the build system was changed to make sure that the native getopt is always used on Darwin platforms.

mirmake is a portable version of the MirOS make(1) program. It also contains the BSD Makefiles, i.e. bsd.prog.mk(5) and friends, some utilities like realpath(1), and a compatibility library (libmirmake) that has implementations for some functions not available on the target system—most notably hash functions for md5, sha1, sha2, and tiger hashes.

mirmake-20080218 can be downloaded from /MirOS/dist/mir/make on this server. The file hashes are:

MD5 (mirmake-20080218.cpio.gz) = 5e6ecc4432b61e39e9907737d058309c
RMD160 (mirmake-20080218.cpio.gz) = 75de9d3a51570948935b31fb8a280a3ad00757a7
SHA1 (mirmake-20080218.cpio.gz) = 17251b1db87ceed85bd82dc52ca7ace7f7c77786

I am making the first steps on my new MacBook Pro – a fine machine. I had had a bug report last year that MirMake does not build on a new Intel Mac running Mac OS 10.5 "Leopard". During 24C3, I had tried unsuccessfully to fix this. Now I finally know the reason.

It is an incompatibility between the getopt.h and getopt_long.c that come with MirMake and the ones in Leopard's libSystem. It seems that Apple has changed some function implementations to be UNIX2003 compatible. The old ones have been preserved for binary compatibility, and the choice is done via "symbol variants", like _getopt@UNIX2003. The upshot is that you need to use the system header.

I devised a fix (use the system getopt(3) in Darwin) but I now I need to figure out how to make a new mirmake release.

In other news, I am working on the talk for FOSDEM 2008 in Bruxelles. Stay tuned for more stuff about build systems (see news).

The problem from the last entry really seems to be a bug in ksh. In mksh, unsetting a non-existing variable gives a return value of 1. If set -e is active, then the script breaks at this point. This behaviour seems to be the same for the other ksh instances such as pdksh. From the mksh(1) manpage:

Unset the named parameters (-v, the default) or functions (-f). The exit status is non-zero if any of the parameters were already unset, zero otherwise.

However, SUSv3 clearly mandates that unset must return a zero exit state in this case. And it seems to be a sane thing actually: in any case, the result is the same afterwards, i.e. the variable is unset. I could care less if it was set before. Btw, both bash and tcsh return 0. zsh returns 1.

I think I will leave this here for tg@'s return. It will be up to him to decide.

Update (20.02.2008 tg@): Read this.

Oh well, today is december 31. Happy new year 2008, everyone!

I fixed a very annoying bug (for me) in the build system for the website. It turns out that unset returns 1 in mksh (but not in bash) when the variable was unset before. However, make breaks as soon as one command fails. Adding a simple || true helps.

As seen in the news section, I will be in Berlin for 24C3. You should come, too. The talks and the atmosphere are great. The train journey was a bit more cumbersome than usual though: our ICE train was replaced by an InterCity. No reservations, not enough space and worst of all: no coffee.

Update (20.02.2008 tg@): Read this.

I finally have the time (and the possibility!) to write a wlog entry. Thorsten has been absent since the end of november. The problem is that many MirOS services, mostly e-mail and the CVS master, were running on herc.mirbsd.org, which is located in tg's apartment. Now that he is not there, herc is also down. This means that mails do not work. I do not have the subscriber list for the mailing lists so I cannot resurrect these.

gecko's server aka anoncvs.mirbsd.org is the new CVS master. It is running Debian GNU/Linux with an older version of CVS. Thus, I had to fiddle a lot with the scripts in CVSROOT. The most obvious effect is that new commits do not have commitids; they get the process group of the commit process instead, which is meaningless and not unique.

However, as said on the news page, development continues, even if MirOS #10 will be pushed back even more. To follow CVS, look at the changelogs or into the cvsweb. anoncvs is running great.

There were two security-related commits this week: one in src/usr.bin/perl, the other in ports/devel/pcre. Please update those two. A binary package to replace perl in base will be available RSN.

Update 20071126: See news entry.

gcc is crap, and I have no idea how to figure out where the cause of today’s herc/rant outage, kern_sysctl + 0x2f6, is, since it seems to have reordered the functions. Damn it.

I managed to get Dr. Pfeffer to go cache hunting with me again, and we didn’t find one (the most easy one), but the other two (at 18:00 and at 19:00 local time, almost exactly). And at 16:00 I had the cache near his home location ☺ This makes, considering not both sites list all caches, a total of 29 (an offset of 3 at OC, and 2 at GC. Now I dropped a TB at an OC-only cache and am still trying to sort out THIS mess ☺

It’s raining now. Shit, considering I left my bike at Dr. Pfeffer’s place yesterday (due to me staying at some other place over night, where we did the caches), and I need it tomorrow (or rather not, if it’s still raining by then).

As usual, no data got lost (at HERC/RANT), and I held the cache in my hand almost immediately.

The svk port is alive and kicking at devel/svk. I have used it for some simple RCS and merging scenarios with my gf's diploma thesis, and it seems to be working very well, better than subversion even. Startup time is very long (up to 10 seconds, even for svk help) but once it gets going, the operations themselves are quicker. And smerge is just very cool: you can merge one branch into another without manually keeping track of branch points and the like. It seems to me that svk can replace CVS's vendor branches easily, something Subversion does not do. Nice.

I should be starting my PhD thesis in november but I am still struggling with the bureaucrats. The problem is that I do not have the diploma itself, i.e. the peace of paper, as the evaluation of the thesis is not finished. So my employer told me to get a "provisional attestation" (?). Now, suddenly they tell me that they cannot take the one I sent them into account because my thesis is not yet noted. Great. Grr.

I tried to install my T-Com Speedport W701V router at my parents' place, only to have it crashing on me. It seems that the thing is stuck in an endless reset loop. The action of mine that brought it into this state was clicking "Step-by-step DSL installation". So much for T-Com (that does not exist as a brand any more anyway, I think) and for our channel motto Technik ist scheiße und funktioniert nicht. How right.

Update 20071231: T-Com exchanged the router without a fuss. Apparently, this is a known problem.

Synergy effects are the best. The geocacheing software Cachewolf now has automatically generated nightly builds thanks to a FreeWRT box, whose developers help testing and spotting all the bugs in mksh, make(1) and the HTML manpage stuff, and all projects get better. Wow.

I feel a little ill at the moment, which quite sucks. Not even bad enough to be unable to work, just… anyway, my bosses sent me to the Doctor.

I have generated a new 5120 bit RSA host key for use on tear (my new server), so that we can continue working on the #10 release timely even though the box is still not ready. (SCSI HBA woes, at the moment. Gosh!)

Benny, despite busy moving, is working on MirPorts again. Austriancoder and spaetzle are helping with FreeWRT. It’s good to know you’re not alone. Users are giving actual feedback. People (e.g. developers of other projects) are helping pro bono. The world is a good place still. Despite…

Working on two OS projects has very nice synergy effects. Today, as yesterday, it affected manual page availability for both projects in a very positive way.

Damn. pax(1) has a bug on FreeWRT on the targets themselves. (And my WL-500g still doesn’t have any network on trunk, this seems to be a different bug.) And I have no idea yet as to why.

When Dr. Pfeffer doesn’t want to go geocaching, I do. They amount to 26 now, although 1 isn’t listed at the commercial site, and 2 aren’t listed at the open site.

I played a little around at work, and MirOS does boot fine, even with networking, in a Xen HVM configuration. Since it was at work, I couldn’t really do much. It was tricky, though. My boss now wants to know if Windows® would also work (I assume so, now I know how).

FreeWRT nor also has online manual pages, now guess whose script is doing the job. Right. I’m especially proud of the fact that this was in no way paid work (and that ekkoBSD was not for void either). I thought, after having hacked on it for quite a bit on the last days, that my other project (I am, after all, the provisional project leader until wbx@ is no longer EAVAIL) could benefit from it as well — I think manpages are a unique selling point in the embedded GNU/Linux distribution market (others have their own USPs… such as working X11…), and I discussed the idea with wbx@ a long time ago already; now it’s there. Of course, the packagers need to make sure themselves that their do-install targets actually provide the manpages, formatted or not, no matter, to the system.

herc crashed today due to a panic. This sucks. No idea why, and I had no time to check either, since the system is live. raid(4) boxen seem to tend to crash more often under low-ram, much-I/O situations. No data was lost, though.

Okay, the boot blocks are MI, the manual pages are split, generated much faster, and I’ve been working on the AES improvements. Today the weather doesn’t make me (or Dr. Pfeffer either) feel like going out, even though there are some interesting new caches in the area.

I’m still waiting for half of the equipment I ordered, but tear can finally be installed now (I’ll just change the internal cabling a bit beforehand though, to make my nerves more at ease).

UPDATE: Changing the cabling wasn’t easily possible, but I cleaned up the inside a little. But RAID initialisation definitively took too long, and afterwards I filled /dev/rsd0b with /dev/arandom at about 80 KiB/s, which is most truly not enough for an U160 SCSI device, not to speak about the U320 devices I also have in there. So I need a new SCSI HBA, which is supported by MirBSD and handles U320 LVD drives. This sucks. And it postpones usage of my new server quite a bit. The devices almost all share IRQ 11, which sucks even more. I gather from Xen/Linux console output that ACPI devices have 256 IRQ lines? This would be the one and only reason for me to adopt it…

I suppose this is the day my best friend moves back into this city, and since it’s due to his girlfriend I guess credit goes to her. Good. Now we can hang around more than the last years.

Today though I was at the others’ place (at cnuke@ and bogus), together with Kabelaffe, and we hacked a little, everybody on his own project, as we used to do in the past as well. I managed to create self-installing i386 boot blocks, integrated the chain boot block and SunOS disklabel, and now everything is ready for MI installboot(8).

I also started to poke around the kernel for using the hardware-aided AES functions modern CPUs (not crypto extension cards!) provide. UVM swapencrypt is going to use this code soon. vnd(4) will use the OCF — crypto(9) — for operations though, so that users can select the actual algorithm used, and we retain backwards compatibility (plus, software blowfish is a little less CPU intensive than software Rijndaël (AES)). I’ll have to write some glue first though, as the APIs differ. And I’m still wondering how to test all of this… I guess tear will have to hold for it prior to going live.

I still did not make any progress with the merged manual pages though. (Making a list which of them is the same between arches.) Although this would be nice to spot gcc/sparc/nroff bugs as well.

I had some conversation with Miod; maybe we should get boot.net back, as bsd.net kernels can’t be debugged (due to having their symbol table stripped, Miod says, but I don’t think it’s entirely stripped from what I see at elf2aout(1)’s code). But I still need some helpful people to e.g. maintain an mksh port inside OpenBSD Ports… and we definitively need a kernel coder inside the MirOS project.

It’s getting cold. Winter is near. Okay, so is Samhain.

Okay, what has happened the last few days in MirWRT-Land? I’m sort of feeling ill, sneezing a lot, coughing some, having to depollute quite often. I wonder… I am taking some vitamins…

My SCA adapters arrived on Tuesday. Almost in the night. Okay. That was a bad review at ePray. Or so I thought. It did not accept that from me without me passing some kind of driving licence-alike test… weirdos. Then I got a “function not available”, went back in the browser history, had to edit the line again because € was changed to &#8634; which was, in total, too long… submitted… rejoiced… or so I thought. They accepted it — finally — but then asked me if I want to add that guy to the list of my preferred sellers. Yeaaah, of course…

Then I worked on MI installboot(8), got done with the sparc part at a Netzladen hacking session (well, more of a hacking-alone session, fukami just chilled and occasionally smoked (grr…) and Dr. Pfeffer came late — although he hacked on CacheWolf and I helped him test some diffs). I think I can now do what wbx@ prodded me for quite strongly on the phone: create a sparc bootable CD on i386 (or even non-MirBSD, for that matter). After bootxx/i386 is converted, I think I’m going to stop working on this until past the release, but it’s good to see that it works and can probably be easy to finish.

On Wednesday, I managed to add the SCSI discs (all five of them, after stealing jumpers from old boxen) to tear. They work quite nice, except the chassis is now humming. Well, 10kRPM drives. Okay. I low-level-formatted them of course, neither do I need or want the data which was on them before nor do I desire bad blocks if I have the chance to scan for them (Benny asks me “why” every time I do low-level formats, but I still believe in them, even though these aren’t MFM or RLL discs any more). And they’re really fast, I was amazed. I think I won’t need to RAID 1 the CVS repository into memory for speed, even if I could now.

Well, next on the agenda would be migrating to tear while releasing and already coding some post-release stuff. Damn, hen-egg problem. I wonder, but not about how I’m going to solve it but when I get my lazy arse moved ☺ We’re making good progress nevertheless.

Benny agreed (via the phone) to speed up the build by several hours (on a P2-350 — probably much more on the sparc) by just linking HTML manual pages if the catman page is only a link too. I implemented it, but since these generated from <bsd.man.mk>’s MLINKS are hardlinked, I needed inode tracking and cacheing, ino_t is unsigned 32-bit, so mksh’s arrays need to be able to use 32, not 31, bit as index. Done now, and it pleased me to shock my packagers, doing as if mksh R32 were impending ☻

Ah yes, Opera got a security upgrade, and while playing with AWStats at work, I realised our Lynx didn’t get counted as Lynx but MSIE… sucks though it helps with certain sites, so I changed that. Can’t have enough people noticing the small ones (Lynx as well as MirBSD). Really.

I am going to have to write an install.sparc(7) page — talking with the MidnightBSD people I realised few people know how to netboot a sparc, and since all of our manual pages are online, I just could have pointed him there. But the mnbsd people are all quite nice.

Today morning, I showed our website to a colleague. She told me she quite likes the design, the colours don’t hurt the eye. This has to be said (she allowed me to quote her here), in the face of all the critique on the current theme. (That said, the theme is not going to stay in an unmodified form though. I recognise there are problems.) She suggested we put the gothic ASCII-Art text into a graphic, together with the two logos. And “IEs for Linux” shows that a certain graphical browser has quite big issues with CSS in general.

My SCA adapters still haven’t arrived. I guess someone cries for a bad review at ePray.

mksh R31d is there, finally. Well, at least the distfile. Since I’m at work right now, I cannot really do a proper release commit series (www, ports, etc).
This release is dedicated to spaetzle@freewrt.org who kept bugging me to look for another busy-loop spinning bug, and to a certain BSD-licenced compiler whose optimiser bug has yet to be fixed, still. (I’m running with /bin/mksh compiled with pcc without -O but with -xdeljumps o MirBSD right now.) Ah, and the distfiles are there now, finally, too.

Dr. Pfeffer now has a statpic (we had a meal (Döner Kebap) together, and I “persuaded” him to configure it) at OpenCaching now, see above, but the cache we wanted to do during lunch break, with my colleague, was unavailable. Closed during the winter, and the cache owner neither knew about it nor believed me at first (though she later fixed the information and thanked me or so, but at first… nah nevermind).

Benny’s back and in France again (ok he’ll travel to Dresden tomorrow, I guess he’s more on the train line than on the internet line ☺). Just good to know he still cares ☻

Waldemar wants me to produce SPARC CDs… *sigh* guess I really have to work harder on the MI installboot(8) stuff. Damn. But o-okay.

Sometimes, cloning autotools isn’t fun. For instance, when you realise that cutting down its triplet of compile-time, link-time and run-time checks is good, but you do need compile-time checks, since not everything can be done as link-time check (e.g. checking for declarations when using deliberately wrong ones to make the compile clash — these clash at link time when the declaration actually was not existent). Oh well. Fixed.

FWIW, current CacheWolf development snapshots are available from my server, please don’t overuse its bandwidth…

I did another two geocaches with Dr. Pfeffer, who doesn’t have any internet access nor a statpic yet though. Update: He has now, it looks like I’ve almost got him there.

mksh R31d is released, I just didn’t have any time to prepare a distfile yet. Please see the list of changes, everybody is strongly advised to upgrade.

Ordering things online sucks. I still do not have my SCSI converters. No progress on the MirBSD server front yet. I hope to be able to host a little box at my workplace. Maybe a VM. And the other equipment I wanted (flashlights for geocacheing, for instance) only has been brought to the packet service now, let’s see how long it takes… I wish I had a beer.

Phew, we got a #10RC7 for both i386 and (days later) sparc now. Now I can refine the manifold-boot setup CD (7 MiB or so), maybe I’ll be able to create it on only one architecture, since “self-installing bootxx we are almost there”, and hardlink the HTML docs which are the same on all architectures (ignoring bugs in nroff(1) on sparc).

But first, let’s go geocacheing again, with Dr. Pfeffer, Addy, gecko2 (alias Turodagnir) and myself. It’s good weather today. And because I’m not going to have tomorrow off I’d better enjoy the weekend.

Today, not only me, but also gecko2 found another two caches. And: I think I hooked up Benny, as I found a couple in the vicinity of his hometown at his parents’ place.

Why does it always have to be us BSDers that have to find the bugs in crappy GNU saftware? Today’s thanks go to Peter Valchev for being truly helpful in tracking down something that I first thought of as an endian difference issue in UMAC-64 (the new one in ssh(1), you know). It turns out this was a bug in libgcc, and while libc contains “better” code, it wasn’t used due to the linkage order. Fixed. Thanks, pvalchev!

We’re a lot closer to release, despite the recent fluctuation, as I’m done with preparing the manifold-boot CD stuff and could quite simplify sparc support overall, as a side result. Now, just minor stuff is still to be done (especially my new server and the SCSI cabling), and we need a few full OS rebuilds… *sigh* so much for sleeping quietly, subsequent nights I will enjoy the sounds of a SPARCstation again.

I wonder where Benny is, he hasn’t been contactable for a while. Last thing I remember is he tried moving to France again. There you go…

The guy from the company where I bought my new server called today. A service to the customer… they always call and see if one is happy. That is new to me (as is that weird chassis) but not bad.

pkgsrc® is hereby officially dead. I cannot write an eMail (or rather I get bounces) to the official MAINTAINER address of mksh, and nobody’s caring. Hey, even Debian does get it! (And others… you should be ashamed of yourself.) Coverity mostly relies on pkgsrc® — otherwise I’d have posted results of their latest scans already.

We’re actually very very close to release. I’m helf off by the issues with the SCSI cabling (see my earlier posting), my laziness (or rather, I’d like to call it being unable to enter deep hacking mode for a while ☺ honestly!), and a few missing things.

Well, I went and did some of the things. I read much source of kernel random(4) and random(9) stuff as well as OpenSSL, fixed a lot of things I found (oO ☻) and wrote a script (mircvs://contrib/code/Snippets/ is a good place for them…) to create a seed file for “openssl genrsa -rand”. This can be used to generate a large (4 Kib or so) secret key; since I still have the sparc and nwt (80486DX-33) around, I’ll just use two keys on tear: this one publically and an 1024 bit key bound only to the LAN (might even make it IPv6 only to annoy my visitors even more…). I also made arandom(4) reseed much more often if we have a hardware RNG (every 64 seconds instead of 10 minutes; I hope that’s okay, but we get 64 Bytes of new entropy every 10 ms, so I guess…).

syslog(3) and friends are stupid, they truncate waay too often; I did a würgaround in logger(1) for now, letting it split up conservatively — at 512 bytes already; doesn’t catch all-binaries but what the hey…

There’s a new src/distrib/tools directory now, with releng’s playground. Like MirOS #7ter, we want to make the release CD dual-arch, bootable. But since #7, things have changed — you can use our ISOs as a kind of Live-HDD, Live-CompactFlash, etc. so this functionality must be retained (I know how now, I just have to code it… and get my sparc PROM to boot off a burned CD-RW). And I could even do this to the #10 series mini-ISO (cdrom10.iso). Having an ISO of approximately 8 MiB size, boot both architectures off it. Yay! (as benz would say)

pcc’s in the works, I like the activity but not the split and some of the directions… but at the moment I’ll only monitor (and contribute the needed MirOS support bits, of course) since we’re busy with other — and much more important — things, the impending release for instance.

herc is in the NTP server pool… so will tear be, but I wonder if we should do a same for entropy distribution (like randomnumbers.info et al. do)? Since the box has a hardware RNG and a pretty good in-kernel mixer, I’m having it exported to the public anyway (but not too well-known since my internet connection cannot handle much load).

Speaking of entropy… ICQ logs conversations proactively, according to ciruZ, so we sort of decided to just send random data to each other. We didn’t set it up yet, but is anyone else interested? Does ICQ close the accounts when they just send to random other accounts? I have some very evil ideas for mksh scripts… *grins*

Why does every friggin’ SPARCstation 20 from the last millennium make it easier to build a SCSI-based system with the modernst SCA hard discs (U-320 LVD) than a brand-new not-even-off-the-shelf server system?

The good part is I learned much more about SCSI today again.
The bad part is: I gave out 1 € (plus 10 € shipping…) on ePray to gain a bunch of 68-pin ./. 80-pin connectors. ☹

The remainder of my new server is looking good, though.

Besides bumping the number of geocaches up, to 18 caches found, three geocoins found, and only one DNF, I admit having been a tad lazy lately.

At the moment, we’re testing the current HEAD code intensively, and I think that, despite theoretically being able to release any time now, a good chance to find bugs (like these which led to mksh R31b, or that my vnd(4) encrypted home works more reliably now) shouldn’t pass unnoticed (and Benny is still fixing ports).

I will probably get my new server tomorrow. It will replace herc, and — sadly — its hostname too, since a Micro-ATX board only comes with two PCI slots (at least one too few for me anyway) and no ISA slots. But it is a fast machine, I already got the dmesg, has one Gig RAM, and at the moment, the Padlock AES/RNG stuff isn’t yet recognised (some hacking to do — but solidly setting it up is what counts now). Well, this opens an opportunity to use some spare Pentium Ⅰ board to write a framebuffer in the kernel for the Hercules card (and compatibles). If I get time.

You wouldn’t believe it: at work, I’m writing a user-friendly (woah!) front-end using dialog (yes, me!) for BIND (me the djbdns user)… and it is still sort of fun (sure, I’m using mksh for both the main program, input loop and all, and parsing the zone file (slightly more restricted format than default, but good enough for us). It runs ssh as coroutine. (That’s why MirPorts’ misc/dialog is more recent than Debian sid…) When my boss sees it, I want to see his face. I bet it’ll be not unlike that of Benny when he first saw the Baselive-CD preboot selection menu.

Is this called summer laziness? I’m actually not really interested in hacking at the moment, even though we could use a release, and I could, as well, make use of my nmeadecode(8) project, since herc is now a pool server for Germany and Europe at the NTP Pool project. Quite a good one with a score of 20.0 at the moment, and an average offset of -53 ms (to the pool tester, which is an SNTP client), with a consistent and rather low standard deviation (0‥-100 ms), in contrast to Hephaistos, whose is pretty unpredictable. And nobody bitched for me using OpenNTPD.

Ah, yes, the dual-arch boot CD issue. I have plans. Stay tuned. Is is possible after all, I just need to get my SPARCstation 20 to boot off a CD-RW to test it…

Spam to the company mail account sucks.

Ah yes, and I better not respond to Benny’s above posting. But I will see if svk can be of use to access svn repositories, since I’m bound to use them in a couple of projects of mine. Well, not my projects, since, obviously, they would be using CVS then, but which I work on.

Benny and XTaran have won, mksh (or rather, dot.mkshrc) now has dirs, pushd, and popd (exactly as csh(1) had them). The hooks chpwd as well as precmd were added to please some zsh freaks. (The dirs stuff uses chpwd now, for simplicity.)

mksh(1) still doesn’t correctly work (continue is broken) with the promising portable C compiler though. It really is a pity, but I don’t have a clue (yet).

(Damn, now I coded a whole lot of crap, even added C7 RNG/AES support despite planning to not hack. I guess I should get a life. On the other hand, the weather isn’t promising enough for that.)

Writing a diploma thesis is really a job where you can see how the UNIX style is superior to what Windows users are used to. Using (in my case) LaTeX and a Revision Control System (Subversion here), you enjoy lots of benefits: I can give the source code to someone for revision, tag the version I gave out, check in what I get back and merge. Compare this to using Word (eek) or OpenOffice.org: you can track changes, but if you work on the document in the meantime, it is not so easy to merge changes from two sources.

Add the fact that the result so far looks absolutely gorgeous, with nice margins, well-placed images and a great text font (I love Palatino). And the LaTeX way of doing automatic numbering and links worked perfectly when I decided in the middle to restructure what I already had. Managing my references with BibTeX, especially with BibDesk as a Mac OS front-end, is also a pleasure.

We had a discussion on IRC (#mirbsd, of course) about Revision Control Systems, triggered by tg@ whining about Subversion again. Let me just say that we have different opinions on this topic.

Apparently, GNU arch is more or less dead, many people are jumping on the git bandwagon. bzr (sp?) and Mercurial (hg) are also popular choices but seen less usable. I will definitely port svk though. They promise the good parts from Subversion without the overhead of libsvn_wc. "In fact, one of the motivations for writing svk was to get rid of that Spaghetti code." Sounds interesting. Working copies are smaller, checkouts are faster, mirroring is easier, they claim. Let's see.

I’ve just put the last polishing steps for some in-tree software into cvs, and I suppose the release is imminent. If but XTaran would deliver some more sparc test results… I have no idea if XFree86® works there, nor if the ISO is bootable, I can just hope.

Since I’m fairly busy with my dayjob and have increased duties and responsibilities at FreeWRT, and Benny is working on his diploma, and we’re okay with the current code quality, I assume that what we have in the tree right now will be the final release.

Oh damn. I still need to work on the dual-arch bootable CD stuff. Crap. That means some more coding (where the assembly part is the easier half…) and possibly testing. But it’ll rock.

Just to not strain the poor packagers, there will be no mksh minor release tomorrow ☺ At least, I hope there’ll be no more late real life bug reports for problems I deem important enough.

I “found” my 10^th Geocache, which means that, on OpenCaching.de, the website I prefer for being devoid of the capitalistic attitude which prevails on GeoCaching.com, despite most other CacheWolf involved seem to not like, use or even know it, I can mark one cache as favoured. I did. Nice view from there…

Woo-hoo! Even homsn got his arse moved, Arch GNU/Linux now has mksh R31c, like Fedora, Gentoo, soon Fink. World domination, anyone?

I am doing pretty much nothing for MirOS these days as I have only three weeks left for my diploma thesis. Hopefully, in a month or so, I will be a real Diplom-Chemiker. We will see what happens then.

I spent a very relaxing weekend at my girlfriend's place. On the way there, I got so bored that I almost finished a research paper ;). Writing a paper is nicer than writing a thesis because you pick one specific subject and try to make a concise statement. It also makes you realise which experiments are still missing ;).

And the paper is in English. I discovered that writing scientific texts in English is easier for me than in German because the majority of existing literature is also in English. After a while, you get used to the typical style, and it sucks when you cannot reproduce that in German. I also have some problems with terms for which I do not know the correct German expression :/.

Question to the interested reader: it seems that if you encounter some totally bogus paper, it is almost always published by Elsevier. Why is that? Discuss.

I have been in Berlin for some kind of business trip, and I finally could get a hold of a beer called “Krusovice”, it’s a black beer, and I really like it. Benny does, too. I drank it in a restaurant called “Ranke 2”, which was recommended to me by Przemysław, thanks a lot! We managed to buy some bottles in a beverage store (alongside with a lot of bottles of “Berliner Faßbrause” — the local speciality, some kind of herb lemonad — as well as a few interesting beers, such as Neuzeller Kloster-Bräu Kirsch Bier, Kirsch Porter Das fruchtige Schwarze, Neuzeller Kloster-Bräu Schwarzer Abt Schwarzbier, Duckstein (Rotblondes Original — Auf Buchenholz gereift). Beats my usual Schlösser Alt with Cola and optionally cherry juice ☺
At Ranke 2, I also drank a “Danziger Goldwasser” to close the stomach after the good diner, which was something new for me as well; I must admit I liked it, but not as much as croatian Julishka.

mksh R31 is out, grab it while it’s hot ☻

The SPARCstation is almost done building just another snapshot, with the bootloader vs bsd.rd bug fixed. I suppose this means that the release is imminent, despite XFree86® (4.5 → 4.7), ncurses (5.5 → 5.6 or even CVS version) and maybe more waiting for updates. Ce la vieh. We’ll do that afterwards, and of course we ensure that #10 users can use later MirPorts snapshots.

Oeps. I had to re-roll the mksh-R31.cpio.gz distfile. Damn. But the Debian upgrade was already done. On the other hand, lintian did bitch a little, so Debian accumulates two “nice to have” fixes now.

I wonder why companies produce such crap. No, really. And I wonder even more why the previous owner (a colleague who went off looking for a better paid job, probably with less good working climate) asked for exactly this laptop, as it doesn’t even work without any troubles on his beloved Deb*an G*U/L*n*x (all product names are changed because I don’t want to be sued). It's a S*ny V*io, a brand which never had any good reputation in our circles. MirOS won’t run natively on it our of the box (okay, I could build a custom kernel, but hey, maybe these dual cores can better be used with VirtualB*x or something like that, even be running in parallel with W*nd*ws). But still, I had my fair share of fight with the D*b*an-Inst*ller, which doesn’t even come with fdisk. I managed to wreck it, expect this wlog entry to be amended by a photo of that (which, at the moment, is located on the mobile phone of my intern, which has a camera built in).

At least I have a use for these G**gle “Go Code!” stickers now — to cover the built-in “webcam”… Who the fuck produces or buys such kind of crap? Sorry for the emphasis, but I just do not fucking get it. And of course, w*ldi and b*nz tell me that there are no such bugs in the Inst*ller mentioned above. (There are a lot more, for instance I cannot choose UTC as timezone, neither grub2 nor grub nor lilo worked — but I got the latter installed manually now, etc.)

This new orking device still has more hours of “fun” for me, but as for tomorrow, I’ll be driving to Berlin. I left that craptop at work. I’ve got better things to do in the evenings.

Diego Biurrun, who was both actually present and interested in my FrOSCon talk, pointed me to a paper called "Recursive Make Considered Harmful" [1]. (Thank you!) In response to the question he asked about this during the Q&A session, I said that recursive Make is automake's default and only behavior. This is not true.

The author of the paper starts with an example project very much like mine (two .c files, one header) and creates an artificial module boundary. Unsurprisingly, dependencies do not work as they should any more. I have witnessed this phenomenon and have seen large recursive build systems fall apart. Continuing a mozilla build for example takes unacceptably long, and you see many IDL files unnecessarily being regenerated.

However, implementing the one-Makefile-per-project technique outlined by Miller is actually very easy to do in automake. Older automake versions did not have good support for source files in other directories but this has not been a problem at least since automake 1.8.

If you paid close attention, the Makefile.am I created in the talk was not really recursive because only one Makefile actually did some compiling. In fact, you can merge Makefile.am and src/Makefile.am into one file, like this:

ACLOCAL_AMFLAGS = -I .

bin_PROGRAMS = hw
hw_SOURCES = src/main.c src/hello.h
hw_LDADD = libhello.la

lib_LTLIBRARIES = libhello.la
libhello_la_SOURCES = src/hello.c src/hello.h
libhello_la_LDFLAGS = -version-info ${LIBHELLO_VER}

If you insert all your targets into one Makefile.am, the automatically inserted stuff from automake only needs to be inserted once, thus you get a lower overhead. Dependency tracking should be even more efficient, and build times should decrease. One technique proposed by Miller is to use per-project Makefile fragments which are included by the main Makefile. Automake supports an include directive, which is directly evaluated during the automake run and thus fully portable.

One problem I see with this approach is that all objects are put into one directory. Thus, you will have problems if different files in different directories have the same name. For example, compiling both foo/main.c and bar/main.c inside the same Makefile.am will not work. I do not know if there is an easy solution to this problem—perhaps an automake option to put the object files in subdirectories. Any automake expert reading this, please contact me if you know of something.

References

1. Miller, P. A., Recursive Make Considered Harmful, AUUGN Journal of AUUG Inc., 19 (1), 14–25 (1998).

   URL: http://miller.emu.id.au/pmiller/books/rmch/

Yesterday morning, wbx@ prodded me to hunt that bsd.rd vs bootloader, bsd.rd.net working, bug, together. Turns out I had to fix an error in a check if the kernel is not too big, turns out bsd.rd was too big, so we decided to bump the RELOC value a little.

Forking is “in”: MirBSD, FreeWRT, Ewe, GForge… what’s next?

Things that suck: too small partitions for mirroring /cvs on a 2 GiB drive which also hosts a #7quater installation, with too big of a block and fragment size to fit all comma-v files.

I’m somehow not in a hacking mood right now, we basically just need a few more people testing #10RC6, then recompiling (from a clean install) it and releasing that. Or so I think.

The weekend is intended for relaxing, I think I have done so. I’m not even in the mood to bike to the city and have some big dinner, for me a couple of butterbread and some sinalco cola+cherry juice+beer will last that evening. The only thing which sucks is that I have to wash clothes because I’m going to Berlin this week (business trip). I really hope my laziness will allow me to get a set of clean clothes for THAT ☺

I wonder where Benny is, this weekend. Probably relaxing, too ☻☺

I had some feedback for my FrOSCon talk from Jörg—better known for MirOS users as J?rg—Schilling, who did not even attend it. Somebody told him I had been mocking his poor excuse for a build system. Since then, mails have been going back and forth. He accused me of spreading lies, being generally misinformed and (that is the part I like best) having made a bad talk. In fact, the latter is also the subject of his mail. Good to know that he can sense what I said from 500 km away.

It was impossible for me to get any kind of factual argument across to Jörg. For instance, every time I said something about his Makefiles, he kept on talking about smake. I am not the slightest bit interested in smake. The really interesting questions were not answered by him: Why are there 325 files in the RULES subdirectory in the cdrtools source (and 48 READMEs)? Why are there two Makefile fragments, i386-openbsd-cc.rul and i386-openbsd-gcc.rul, which are otherwise completely equal and choose to ignore the user's choice of C compiler? How do you want to achieve platform-independence if all you have is 325 platform-specific files?

The answer is that you should just install smake, and thanks to its magical capabilities (which J.S. calls "automake functions" to further confuse everyone), everything will work. I refuse to look at smake, but for me, this just shifts the problem further: How do the Makefiles of smake work? Should I install smake to build that too? Keep in mind that this is somehow better than automake, which works with the native make(1) or GNU make, and for which no additional software needs to be installed, at least a priori.

So, to conclude, I stand fully behind what I said, and I will not "correct" any statements I made. Regarding the licence issues behind the Schily Makefiles, different people have different opinions. And Jörg, you should be ashamed of yourself for saying:

"Schade, daß Du die Lügen der OpenSoure [sic!] Feinde bei Debian unreflektiert wiederholst."

I mean, really.

For its birthday, it got further improvements to the www source code as well as a merged page more; unity of mind, and consens between developers regarding how to handle the next events and J�rg Schilling, and nobody, I think, noticed. But how fitting that Theo de Raadt is now guilty of doing actual harm to his own project, by shouting at Linux® developers, so that they use GPL-only instead of BSD/GPL dual licencing for their fork of his ath(4) OpenHAL (after fixing a licence violation that they were correctly accused of). At least me and nbd from OpenWrt think so.

These are quiet times, and I worked long today (biking up a mountain to the server housing farm *sigh*, and hating Subversion even more), as did, or so I think, Benny. My SPARCstation (demo.mirbsd.org) finished creating a new snapshot, hereby dubbed MirOS #10 RC6, which will be uploaded soon, along with a recent one for the i386 architecture. I hope it works really well, so that we can release as soon as possible (and tested; XTaran took it okay and is currently backing up his stuff, thanks).

FrOSCon is over for me now as well, with the end of the two days I took off to replenish after the event. I used it to chill, rest, not do much — so far it worked, I feel okay going to work tomorrow, not strained.

Since yesterday, I’m officially a developer of Cachewolf, thar software Dr. Pfeffer and I use for geocacheing. And I fixed a bug. In Java™. Urgh. But it works now (for some low values of work). And they use svn. Yuck. I can’t use non-7bit chars in the commit messages, because J�rg Schilling’s fucking BerliOS setup wrecks the emails if I enable use of locales in the post-commit hook, which worked for FreeWRT. Line endings as well as character encodings (they use mostly windows-1252) are cruel too — but probably can be fixed. I need to start some discussion about that. To use gcj for compiling, I’ve prepared a comp-gcj.sh and even a simple Makefile, but I want to clean the mess up some more before committing the nuggets, because it’s easier (also for Ewe and Jewel) to have the classes from external libraries unpacked in the tree.

GNU ed sucks — it is less BSD ed(1) compatible in more recent versions… I bet they’ll be using GPLv3 next — and GFDL for the man er… texinfo pages (now what, GNU software doesn’t use manual pages any more, and almost no GNU/Linux distribution has texinfo).

I would have listened to the grml talk at MoeLUG tonight, but nobody of my mates with cars wanted to go. Too bad, I really liked the Altbier.

Along with this post, I am inserting a vim modeline into the source of this page. Finally.

As said in the title, FrOSCon 2007 is over. IMHO, it was one of the best conferences I have ever attended. I loved the relaxed atmosphere, the cool people, the catering, and the Free Beer—sponsored by O'Reilly, who really deserve more credit for that. There were many niceties and little things that made the conference truly great—like the Wii lounge, where I played some tennis against Marius from AllBSD.

I was not at all happy with the way our booth was handled though: It consisted of a row of tables, the showcase computer standing with the monitor to the front. Behind the tables, three to six persons were sitting crouched behind their laptops, only one of which (tg@) actually using MirOS. The others were not even familiar with the project, with its goals or with the content of the flyers. The showcase was almost only ever used for showing something when I was present, doing so myself. Flyers and CDs were not proactively handed out to visitors, they had to get these themselves. Note: I will not accept this behavior again from someone with an exhibitor's pass for MirOS.

On a more positive note, my talk automake, autoconf, libtool—an introduction to the GNU build system was a success. The lecture hall was almost full; the audience was interested and critical, asking many questions along the way and at the end. (Thank you for that.) I started 15 minutes late because the video team did not have a new cassette for recording. They called a helper over DECT, asking for a new one. The helper came five minutes later, asked if we were the ones that called and went off again to actually get the cassette. It was truly a WTF moment for all of us, the audience waiting. I made it through at an impressive speed: All my original slides as well as the demo were in, and even with questions, I made it in about 50 to 55 minutes.

In the abstract, I had made a bold statement about cmake being a flawed concept compared to the autotools, but I did not have any data points to actually back up my claim, and promptly I got some questions about this weak point of mine. Sorry for the unsatisfying answers.

The slides in PDF format as well as the code examples are downloadable from the FrOSCon program website. If you attended the talk, I would be happy to receive some feedback, remarks and criticism via the form on the same page.

Oh, and it is true that the website here is somewhat broken now because I did not commit all of my changes. The point is that commit windows, with a r/w CVS repository, are rare now for some reason. Halfway through my commits, I was distracted and could not finish. By and large, and contrary to popular belief, a conference and exhibition where you have a booth is not a good place or time for hacking.

Where’s my MirOS flag, I wonder? Anyway, we finally got ISOs, I am fully under strain, and the food is good and the net working.

ciruZ did me a mcabber port, now I need to learn how to use it. Dropping all non-Jabber connections was fun. (He tells me to write “js” as name, as he doesn’t use that nick any longer. I guess you call it “Richtigstellung” in German press law.)

Benny is constantly annoying me too… I’d better go and eat something.

… the grill is dimensioned much too low and it’s hard to get the grilled sausages. Luckily, users contributed some Altbier, and other visitors went to the barbecue queue for me. Thanks!

Something seems to be wrong with manpages on the sparc. Maybe that could be fixed during the next build. No idea. I guess it could be a caste-align issue. I do not even want to fix the bootloder, the code is so ugly.

This is day 0 of FrOSCon 2007, the event whose planning of a “LinuxTag Bonn” started out during some car ride to FOSDEM. I’m not yet there, because I am committed to committing all the local Ms from my various source trees first, as “herc” (or rather rant) is a little working right now (after rebuilding the RAID several times during last night, which made me sleepy…) and I’d rather get it done now.

Expect a wlog entry per day per developer (well, at least me), and these are constantly updated (maybe, we’re lazy slackers after all).

*sigh* Benny updated the index but didn't commit the other migrated HTML files. This website just became more broken in a sudden.

Interesting: geocacheing now works, but CacheWolf cannot handle 4800 bps — but cu -l /dev/cuaU0 -s 4800 <>/dev/ptyp5 >&0 plus using /dev/ttyp5 as “serial” device helped…

We just ate a little and now are going home. FrOSCon looks superb, but a Live CD Image is still not there. It's closing time...

This sucks: we apparently also got the wrong cable for the Zaurus spitz (SL-C3200) — one that does not support USB host mode. Add the Ewe runtime issues (now: segfault in GetLocale() or something like that) to it — plus the CacheWolf issues… damn I think it would’ve had some great educational impacts on our geocacheing newbie of the day. But odem was enough…

I finally got my GPS mouse, wbx@’s is an uslcom(4) at 9k6,8N1; and mine is an uplcom(4) at 4k8,8N1; no problems so far, this is actually the very baud rate specified by NMEA 0183, and its precision is better by one post comma digit (unsure how to call it in English). This means I’ve been outdoors again, with quite some success. I suppose I’ve got to fix the Zaurus issues with Ewe soon — but Cachewolf seems to have some trouble with my version of the GPS mouse which transmits the GPGSV lines only every second second (heh…). In short this means it didn’t work and we were left with wbx@’s less precise mouse which, interestingly, had fairly good precision this time but didn’t seem to actualise the position as we drove or walked. Weird. Or CacheWolf is a little bit more buggy than we thought.

Looks like I’ve got to work on nmeadecode(8) now for real. If only…

mksh(1) now got a “rename” builtin just calling rename(2). I fragged an embedded system (Mikrotik Routerboard 532 with FreeWRT) by trying to move the old directories aside and new directories in place during an upgrade, not remembering that /bin/mv was dynamically linked…

I could reproduce the “detached mksh busy loop” issue on the Zaurus but not track it down (of course). So far, it seems to only happen on Linux — I definitively need help from the community here.

We got MirOS #10 RC4 (i386) out as well as #10 RC2a (sparc), now with X but still not really tested. Both on the mirrors. After reïnstalling that SPARCstation and fixing some SIGBUS, I’m now trying to get a final candid for sparc out before actually releasing #10 into the wild.

*sighs* I guess I’ve got to add -Wcast-align to COPTS globally on my sparc, do a full build’n’fix round. Developers just aren’t aware of these issues, and gcc on i386 can’t even reproduce these warnings with an extra option (say -fstrict-align, that’d be great).

Benny (who seems to be much too lazy to do any wlogging) has to cut off way too much content from his talk. He said he needed 1:06 hrs and didn’t even mention everything… too bad, it really was great stuff.

My SPARC is building XFree86® for the umptieth time, but I still expect it to finish cleanly this time. We nuked all the cropped fonts, will only install the full Unicode fonts now (needs some more RAM, but less overall as you need some of them anyway, and reduces build time); on very low-mem systems you would want to build your own font subset anyway.

I invited the XFree86® developers to FrOSCon — all two of them — but it looks like they can’t come. Too bad.

I started writing “I’m” instead of “I'm” — typographically better, much better in fixed-misc font even, but drives people crazy (and away from OS without Unicode support such as OpenBSD and most Linuxes), funny. Oh, and swishy, our IRC Kiwi, is using my keyboard layout now too. Strike!

With MirOS BSD #10 we’ll release mksh R31, no big deal. It needs tests, just like the sparc port, so please check it out of CVS and test it.

No progress on the nmeadecode(8) front (yet), but fixed some endianness bugs in OpenNTPD. I should call it MirNTPD, I suppose. And rdate(8).

We also spotted some bugs in the ports infrastructure again (better pre than post release ☺) and I fixed the problem that COPTS were passed twice in imake(1) build environments.

Finally weekend, I’m so happy. Benny today finished the slides for that autotools presentation he’s going to hold at FrOSCon, 42 in total (wonder why…) and is happy about the weekend too. Open source developers should — in my opinion — get two hours a day off work to hack. (But I can’t really complain, I’m off quite well with my company. Thanks to my bosses, if you should happen to read this.)

Benny is also trying to upgrade the www/firesomething (Iceweasel) port. Apparently with success. I still don’t like applications that use glib2 — even though I probably will need one, as I might be switching to mcabber, from centericq; I dropped all ICQ/Yahoo!/MSN contacts tonight.

Oh yeah. FreeWRT. Zaurus/Ångström. Kernel modules. Packages. Linux JUST SUCKS, get it. Things I take for granted just aren’t in there yet — worse is that most of them won’t make it in since it’s a bunch of packages from many authors, contributed with “here’s my code, please add it, have fun — and someone should write the docs”. Sometimes, Theo is right. He’s also a lot of right (can I say that?) regarding open source’d (what a word…) NIC drivers from vendors, compared with getting specs and hardware and coding one’s own drivers. Vendors suck. More than Theo, at least.

Compared to other operating systems out there, we still suck least.

There are days you just want to go back to bed. This is one of them and I’m beginning to hate my 42½ hour working weeks again.

Lynx is buggy, trying to convery my nicely working solutions to kludges around broken GnuTLS (with no reason to exist), since the latter does not support even some basic OpenSSL function calls, and they have to autofool for it. Guess who’s going to fix that? Okay, I already did, so…

Due to my recent ability to compile CacheWolf with gcj, it’s not in base again. Yeah, call me selfish, it is nothing new. Anyway, it won’t hurt since it’ll reside in the “ada” set which almost nobody installs, and isn’t selected by default either. Also, it’s i386 only (quelle surprise). Maybe someone can make use of it. gcj’s also enough for university students’ needs (except that gij is broken). I recall Benny telling me that boehm-gc with threads wo’t work; gcj uses it and seems to work relatively stable. Java™ still sucks, especially if you have two different line ending conventions in a windows-1252 encoded file and the docs are in the same charset except for some UTF-8+BOM…

Today was quite the nice day. After sleeping long I started my day with style: chilly, having a coffee, and booting my cell phone, which I turned off during the night because I phoned the battery empty last evening… wbx called me, came over later (after some mksh(1) hackery: set -o arc4random was added), and we went geocacheing again. And hacking. The SPARC build’s complete, Ewe (that single-threaded embedded VM for Java™ which CacheWolf uses, that’s the application Dr. Pfeffer likes to use for geocacheing and I’d even use as a GPS navigator…) works natively on MirBSD — what more do I want? (Ok, I could think of some things, but I’m happy nevertheless.) I face bad future though: we infected wbx@ with the gc virus, so he’ll make me port X11 to FreeWRT to use it on the OpenMoko with his nokernel target I guess. *sigh*
Oh well, I can use his Zaurus in exchange then; that beast at least has longer-lasting batteries than my notebook.

I've uploaded the snapshot to /MirOS/current/sparc/ (the older one is archived at sparc9), but did not announce it yet, because we need to have some more persons actually test it and report success for especially the vt100 wscons change. The floppy image is probably going to die though, and installboot doesn’t like the miniroot (yet).

I didn't get to hack today, plus I slept too much (06:00 ‥ 18:00), thus missed some of the day (did not expect it to be sunny tho) and was tired. To add insult to injury, when having diner in the city some freak thought it would be nice to flatten my bike tyre. I had to walk back, and arrived home at about midnight. I guess it's reading idly for tonight, maybe that hacking mood will get back some other day. I've discussed nmeadecode — or rather the idea for it — with some friends now and like it more and more, as most of them do. Still have to discuss it with Pfeffer.

I just built gcj on odem (still works, yay), took much over an hour, at 1 GHz even, and will try to build CacheWolf with it (can become developer then) to create a jar (for sunjdk or so, just to verify) or compile a jar (official one, or one built myself if possible) to a native executable to use instead of ewe with compat_linux(8). If that works, I'll build gcj on wbx@'s Zaurus, even if it takes a week or so. Cross fingers for me.

Seems as if ciruZ likes the geocacheing concept. Oh wonder…

While I didn't have much time today during work due to some lecture, my SPARCstation 20 was busy building a snapshot (and still is, and even will still be the next day or more), Dr. Pfeffer improved CacheWolf further, I bought a GPS mouse at ePray (together with Addy, the “& friends” part from the last try, and an intern who seems to like the idea), read raw NMEA 0183 output from wbx@’s device and conjured nmeadecode(8) which, until we get nmea(4) and friends, will just read the serial stuff from userland, timestamp it, act as simple NTP server (to use with serverloop in ntpd.conf(5) and bound on a different port), and display position data to stdout.

There are interesting things to be found in anybody's source tree. This time, it is Schily again. Well, it could be cleaned up, and arc4random(3) got improved as well, except I still think it should have a kind of timer for smaller entropy adds (⅛ of the arc4 key size).

On the other hand SPARC looks rock solid now, I could even build a gnat (GCC's Ada part) today. Let's hope we'll have a snapshot in 2-3 days, and maybe MirOS #10/sparc will be in #10semel this time.

The website… some things are difficult to do until you go to the toilet and come back with a great idea. Put back mtimes on the pages.

At work, I hacked some major C code for FreeWRT today, namely in the nfotiser (and here[link removed]). That's a complicated program, difficult to do even for me (mostly since it grows high-level stuff, data structures, and not only close-to-machine stuff, especially the SQL stuff), but it turns out to become fun once you get into “hacking mode”.

Tonight's after-work event is LUUSA (at the Netzladen) again.

Waldemar brought his Zaurus, but there's no Ewe VM. On the other hand — after some kernel patches and downloading Ewe149-Linux-x86.zip as well as cachewolf_linux_0.9n_patch_4.zip — the first from ewesoft.com, the second from cachewolf.de — and unpacking, it worked out of the box; remember the device name is relative to /dev (I had to use “cuaU0” and “9600” as baud rate, add myself to group dialer too). I did need some shared lib from Debian GNU/Linux for that (libgcc_s.so.1) though. Compiling Ewe is — at the moment — apparently impossible, it always segfaults both on MirBSD (i386) and Ångström Linux (armv5tel/Zaurus). I'm importing caches from my site (opencaching.de) now, after recentering to my current location (as I am too lazy to figure out the GPS position of my home right now). Now, if I had a huge battery on this laptop (the current one can hold an hour or so only), I could use it for geocacheing.

UPDATE: This even led to spotting bugs in MirPorts and CacheWolf.

It looks as if I finally have found a few cases of bad programming that caused the segfaults and other misbehaviour on sparc. Now let's just come all the former Debian GNU/Linux/sparc users. Okay, I've got to compile it a few times more, and pre-#10 I'd like to do one more code change, namely the switch from suncons to vt100, but… what the hey ☺ I'm happy.

I'm currently building on a 75 MHz SuperSPARC and would be glad if some nice person would send me one of these 233 MHz HyperSPARC modules that've been made by ROSS (I think), and some more RAM for a SPARCstation 20 UP — we don't support SMP on sparc either. And I desparately need 1-2 testers, since I don't use a monitor/keyboard/mouse zs console but “merely” a real DEC VT420 on ttya.

On HTML: see my rants about standards.

Anyway, despite the bugs found during the sparc compiling progress, our tree is in a pretty sane state and ready for release every day or so. I'm postponing it a bit in the hope less bugs will be found post-release, and Benny is still continuing his ports work, despite having to create slides for that talk at FrOSCon and doing his diploma stuff all day. Thanks Benny what would MirOS be without you.

I've still got my dayjob, and working with parsing, databases, SQL, and C string operations at the same time isn't really fun. But it's me who is able to make FreeWRT 1.1 shine, so I've got to do it. I bet it'll have been a lot of fun once I'm done. Just the way there is hard. So be it. On the contrary, my boss's got a Zaurus, and Dr. Pfeffer's geocacheing software (in Java™) runs on it. And he will not only make these trips with us but also lend it to one of us if he has no time but we want to go. That's the benefit of a small company.

The BOSng Duo has found another cache, although with minor injuries, we hooked up a friend to the fun and had a definitively nice view.

Yesterday, I talked some with the MidnightBSD people. I think that they are quite okay and invited them over for FrOSCon. Maybe at least one from that bunch comes, then we'll finally have all BSDs at once.

My dayjob is really keeping me busy, I'm almost only home for sleeping; but that's okay, I even get to do mksh stuff at work since it's the shell of choice in FreeWRT Trunk (to be 1.1), which is what we use and develop. Speaking of mksh, I'd like to release R30 rather sooner than later.

Benny's busy coding WinAPI at work, poor lad, but it could be worse, as STL or MFC or so. On the other hand, he does still fix ports and package tools (we have a security updated BIND now, for instance). Thanks Benny, we really appreciate that.

Speaking of MirPorts — the MidnightBSD people have “mports”, which seem to be FreeBSD ports based (*grins* I wonder why they didn't use pkgsrc®…) and I suggested them to use MirPorts as a second choice (like pkgsrc® for us), after Archite (coming from OgreBSD) couldn't persuade them to choose only MirPorts he'll probably look into that now. We'll help, but we can't maintain it. I just wish dfjoerg would commit the MirBSD bits in pkgsrc®, since otherwise I wonder why I did all that work if nobody's going to use it due to missing upstream inclusion. Lazy git. mksh's still at R29f.

From time to time, I'm uploading a new snapshot to the netinstall dir — it pays off to look there. Today's one was tested on a WRAP board.

I think one of the two HDDs in herc which compose the RAID 1 (software, using raid(4) — RAIDframe) is definitively failing. I need a replacement; maybe I'll do as few activity as possible on /MirOS/ and the CVS repository in the meantime to extend its life span. This sucks, and I see I won't be able to relax next weekend either.

We have a new snapshot — MirOS #10 RC2 — yet not announced on the front page because I need to rsync it to the mirrors first (RAID issue) and put up a torrent (time issue, I do have a dayjob now). But it should appear @ the http mirrors (heph and allbsd.org) soon, in /MirOS/current/, as usual.

Not the best day for fate… herc had an almost-panic in the UFS code, my bedside lamp got crashed by the telephone receiver when gecko2 called me, I had to nuke tcpdump(8) due to §202c StGB, and mksh R29g was only almost tested before release (it defines _LARGEFILE_SOURCE=1 even if it is unnecessary). I am even thinking of removing all pre-R30 stuff off the mksh page once R30 is out (which might, I must admit, not happen before MirOS #10, but it will rock). At least MirOS #10RC2 is going to happen soon, despite some bugs in the pak- kage tools bsiegert@ is trying to hunt down right now.

À propos gecko2: as he (and zeno) are booth staff, bsiegert@ and I have agreed to enable them to post wlogs here, so don't be shocked. kimnotyze, probably hoping to get an entry ticket from me, said he'll help out staff at FrOSCon as well, but doesn't wlog as of now (this may change if he's involved more).

I added cprng to the base system, a daemon tjat tries to gather entropy from measuring the difference between the CPU clock and timer interrupt — an interesting idea, yet it doesn't work on all hardware. Still better to have it than to not have good local entropy, besides it's fed to the kernel random(9) subsystem alongside with tty(4), disc and network IO (and possibly truly random stuff from the VIA C3/C7 RNG, hifn(4) cards or even from hotbits/fourmilab/randomnumbers.info). The mix is what makes it good, but arc4random(3) with multiple concurrent accesses is a difference to more popular OSes ☺ which makes it truly great. So just use one source of entropy — /dev/arandom — which is rather high-quality even if its absolute entropy count is less than the one from urandom.

Somehow, I also managed to wreck something in my other pet project — FreeWRT. It is often annoying to work with Linux, and this shows it one more time: module loading doesn't work, and we have issues with binary-only stuff (BLOBs) too. Still it is slowly being undermined by addition of mir-tools to the system ☻ and — while a̲l̲l̲ embedded system development kits suck, this one sucks the least.

A few days ago, I had access to a Soekris vpn1411 card — hifn(4) — that just worked™ out of the box on a bsd.rd kernel. Nice. I noted that rdate, as well as uuencode(1) and uudecode(1) and their base64 variants were not available, so I added them; rdate(8) even fits on the small floppy (8KiB) and is now The Official ping6(8) Replacement™ on floppy, since the latter is about 28K.

Today I officially declare that gdb(1) isn't usable on MirOS. I tracked down an GCC bug but was confused because a “load register with immediate” opcode appeared to change memory (the stack frame — the print command did not seem to be affected). But hey, I eventually found the bug. Now, all I wait for is Sun to release their compiler under MIT licence… as if.

As part of the GNOME update to 2.18, I added a python flavour to libxml2. This seems to confuse pkg_add and triggers a bug (one more) in the dependency handling code. Incidentally, the code in question is written in Perl by Marc Espie.

To completely fix the bug requires removing pkg(1) completely, which I will not be able to do before MirOS #10. As a first step, the verbose output of pkg is now more meaningful, allowing (hopefully) to debug the issue.

The way dependencies are handled is deeply flawed and is one of the biggest weaknesses in the current MirPorts framework. pkg(1) converts the dependencies into static @pkgdep lines of the form "@pkgdep libxml-2.6.14-2". If a libspec cannot be resolved, the adding almost certainly fails *sigh*.

It's mksh-bash compatibility week, apparently. Now, ${foo:2:3} works, but these must be numbers, not arithmetic expressions, as of yet — this should be fixed soon though. Also, ${foo/bar/baz} (regexp), as well as its companion ${foo//bar/baz} (like g appended) won't be much of a problem but I'll implement ${foo:S/bar/baz/g} first followed by ${foo:C/bar/baz/gx}, make(1)-style (x = to use POSIX extended regexps instead of basic). Of course, this'll only work if a few functions like regcomp(3) are in libc and functional and MKSH_SMALL isn't defined ☺

MirBSD Secure Shell (sshd) is now built as position-independent code to make it even more than OpenBSD's OpenSSH. Yay! (bsiegert@ thinks so too.) Eat this ☻

Before I forget to write this down… today, I released a piece of code I call libnointl. This was an original idea (and implementation¹) of mine, but I only got it because of swishy (Dale Anderson) who complained about his cross-compiling issues and troubles to me in IRC (I'm feeling the same when working for and with FreeWRT; he's also using µClibc, but non-POSIX FSH paths). Thanks for the intuition; if I get famous (FreeWRT and OpenWrt are already adopting it), you shall become it too.

¹) The API is part from gettext's libintl.h header — only from the non-copyrightable interface part only — and the ABI is re-engineered, using nm(1). There is one function actually doing something inside, whose implementation is from Dr. Dobb's Journal (but easily doable within a few minutes, and found at many other places on the net), optimised by me, and adjusted to match GNU gettext libintl's behavious, of course.

Han Boetes suggested to feed it back to the GNU gettext people… I guess they won't like the licence (MirBSD licence + advertising clause, but GPL compatible), as it's a GNU project they probably want the paperwork.

We now have an uppercase Eszett (ſz ligature) ẞ in the fixed-misc fonts and my keymap; it maps towlower(L'ẞ' == 0x1E9E) = L'ß' == 0x00DF, whose upper-case mapping stays at L"SS" though. I sent the changes in the font to Markus Kuhn, hoping he will integrate them in the next stock XFree86® (and that x.org crap too…) version.

mksh is now a better /bin/sh for some people in Debilian who don't know of Korn Shell tradition. Also, it won't try to execute ELF, a.out, COFF, gzip and NE any longer.

Been working on FreeWRT all the time (even during the dayjob), so I didn't have the energy left to do more on the SPARC issues.

Okay, since this thing is here anyway, let's use it to update you people on the progress of the sparc port.

The kernel boots and runs fine. The hardware seems to be okay, the new HDD (sd0) gets a little warm at times, but that doesn't seem to cause any problems.

I reliably get segmentation faults for a lot of things — usually things that access the network, more rarely during compiling. Can't get stack traces as the stack is corrupt.

Just to be safe, I'll try out the other 9 GB HDD tyler got me. Maybe this one is the cause of the troubles, but I don't think so. On the other hand, I might just try to upgrade the existing #9 snapshot to #10 via source and with NO_ADA=Yes set. If that works…

As I said, having sparc support is not a release criterium for #10, but it'd still be nice to have. I already spotted bugs… but don't expect me to hack much on it, I've got a dayjob requiring full attention and hacking skills (mostly embedded GNU/Linux for mips, though, but luckily most shell scripting I do is in mksh, and I can use jupp; no BSD make there though).

Oops, I stayed out of bed for too long, again. Anyway, I had days of illness and enough hacking mood to do something. NetBSD® pkgsrc® works on MirOS now, and I ported over their svis(3) and OpenBSDs sdiff(1). I just wonder why I always have to fix their bugs…

Nah, I have no intent in reviving this wlog, just ranting once.

I should write more often. That said, I am in the middle of the period of my diploma examinations. So far, everything went really well—in the TOEIC (Test of English for International Communication) for example, I scored 985 out of 990.

Some MirOS-related news: I actually added a feature to MirLibtool: code for treating -Wl,--library-after, our equivalent to -idirafter for include files. I was surprised that libtool itself is quite readable code. One disadvantage of my patch: The library-after parameter is written into the .la file; older libtools trying to link against the file might barf with an "unhandled library argument" error message. This is not an issue inside MirPorts but could be an obstacle against including it upstream.

I am currently doing a bulk build of all packages on schaaf (a hopelessly slow 667MHz VIA C3). Doing a bulk build is very easy. You need to do the following things:

1. Start on a freshly installed MirOS system with plenty of space on the hard disk. Set up MirPorts via make setup.
2. Set some variables we need via export BIN_PACKAGES=Yes TRUST_PACKAGES=Yes BULK=Yes BATCH=Yes.
3. Go to /usr/ports and create a Logs directory.
4. In /usr/ports, run the command mmake -k package clean 2>&1 | perl /usr/ports/infrastructure/scripts/portslogger /usr/ports/Logs.
5. Wait a few days for the build to be over. Note that make -k is needed to skip ports that don't build. Some of those will have a bulk cookie and a package file that is 0 bytes.
6. Now comes the fun (?) part: check the build log of all packages that failed to build. Start with the packages with 0 bytes.

Update: I had to cancel the bulk build as the machine was running too hot.

Yay! Benny lives!

This wlog entry was written on the flight from Dresden to Stuttgart. Flying is great: While it would take me about eight hours using so-called high-speed trains, the flight is only 50 minutes!

So we are in 2007 now. 23c3 is over. It was fun, even if tg@ did not manage to come. My security-related highlight was the talk "A not so smart card", where it was shown how trivial it is to duplicate a Swiss PostCard or a French Carte Bleue so that your clone will accept any PIN (they called it a "yes card"). As a user of a Carte Bleue, I frankly found it very unsettling. At least they don't use 384-bit RSA any more, as the Swiss still do. All in all, I attended a lot of talks. I also fixed a few bugs in the pkgtools and ported Gtk+ 2.10 along with dependencies. There are lots of circular dependencies lurking in the shadows there.

Great thanks to Pascal Bleser from FOSDEM for building SuSE packages of mksh(1). Benny already says, in a few years people won't talk about me as "the MirBSD man" any more but as "the mksh developer". And he kind of likes the shell, he admits, even though he prefers that Plan 9 shell called rc. He says the other *nix shells aren't really in active development anyway and only add bload (for example zmodload). I do invest quite some time in mksh, and it stays a scripting shell, but with good interactive usability.

And a rant: X11 doesn't do everything right. Even that old venerable Windows® 3.1 did do mouse-keyboard event serialisation right. When the system is under load and I type into an xterm and press the middle mouse button in between, the mouse lags. Seriously, quite a lot.

An update to the wlog entry of two days ago: I found out that gcc is not entirely consistently with the const qualifier and a means to cast it away: define a union with two slots – a const void * and a void * one… maybe required for const-correctness, because some of the interfaces (such as execve(2)) suck, but it is NOT an excuse for bad code, so don't you ever use it unless impossible AND guaranteed.

I did upload a St. Martin snapshot to unixforge (you can download it here), but there isn't a changelog yet, that's why it wasn't announced. Just so you know.

I spent eight hours in the train today, one more than planned. I had 20 minutes for changing trains but mine was 22 minutes late. Tough luck.

I was going to rewrite the dependency handling stuff in the pkgtools but I had to clean up the convoluted mess that is pkg_do() in pkg_add. The function contained more than 500 lines of code. I did some refactoring: the automatic loading of dependencies is now done in separate functions. I think recursive autoloading via ftp is broken but we don't have binary packages on ftp anyway. However, recording of dependencies is now handled much more sanely. Before, the tools did two passes over the plist: one for checking the dependencies and one for recording them. Wildcards were newly expanded each time so you could not be sure you record the dependency on the same package you checked earlier.

For the user, there should be no functional change. pkg_add now creates a $LOCALBASE/db/pkg/pkgname/+DEPENDS file, which contains the correct forward dependencies for the package. This file is essentially unused at the moment but will be more important later.

These changes will eventually lead to abandoning the dependency checker in perl. I already got rid of all calls to pkg dependencies check in bsd.port.mk; these can be replaced by calling pkg_info -qe.

(Not commenting on mksh(1) changes here, see the ML.)

Yay, my DSL provider just gave me a static IPv4 address. Too bad, it does cost me a lot! 4.90 € per month plus once to set it up. How? IP addresses must not be paid for, they say. And I don't get to set my reverse DNS. Now, funnily, even their own MTA won't accept mail from me, which it did before, from the dial-up pool addresses.

I'm not allowed to be the first one. FreeBSD® Ports beat me by a mere 16 hours, porting frozen-bubble 2. NetBSD® pkgsrc® hasn't updated from version 1, and OpenBSD ports don't even know what that is (probably they don't read news? I don't care though, as MirPorts work on that too). It doesn't play back the demos (signal 27, SIGPROF, during shelling out to bzcat), and once perl crashed with signal 6 (SIGABRT) for me, but that might just be an overly large memory eater…

On the low-tech side, I fixed more stuff in nroff(1). I like it.

While building with MALLOC_OPTIONS=AFGHPZ I found a bug in perl(1)'s regexp code. While caught using the 'P' option (cf. malloc.conf(5)), I tracked it down to an out-of-bounds write to an array of uint32_ts due to a false estimate of the array size. I however can't seem to fix it. Also, and this sucks, I decided to write iconv(1) and locale(1) but as of now, after trying to hack perl and drinking some Met, can't get my, admittedly lazy, arse to do it.

On the other hand, I fixed more in cap_mkdb(1) and hope that OpenBSD will merge the fixes.

I'm also intending to do a quick utf-8 hack to mksh(1) – utf-8 codeset will make backspace remove entire multibyte sequences. Also, I plan on adding a '-c <charset>' or so flag to the editors in the base system (i.e. ed(1) and vi(1)) so that it'll setlocale(3), then if nl_langinfo(3) CODESET replies something different iconv(3) between em (need to take latin1 fallback for 'C' locale into account, discuss the issue on mailing list first).

bsiegert@ has phone, yet no 'net, at home now. Yay!

Actually I won't implement the conversion routine for the editors as I found that it's much easier to just open a pipe to iconv(1), which I added (written myself) today. locale(1) was taken from TNF though.

Note to self: read operator(7) throughoutly, especially the priority level of the ?: ternary operator.

(statpics moved here)

The BOSng Duo has struck again, old fun in a new location, and I found my second cache rather easily.

Other than that, I guess we have a new snapshot, and frozen-bubble 2 with multiplayer is said to be out. I'll check that ASAP. Update: It's true, but IMHO, the new dependencies suck a little.

I'm back, and we have libncursesw-5.5 as libcurses replacement in da tree. One of my teeth still aches, but the doctor will see after it on tuesday and pull the strings from the other wounds. (Yuck.)

Let's see when I can tend to the other projects… (someone want to take on wsconfig(8)?)

Today, I recompiled screen and mc, and made centericq (sort of) utf8 capable… enough for what I need.

After the odds and ends from last time, now the Health End: I'll not be available/online for a few days – I have an appointment at my dentist (ouch!) tomorrow, for 1½ hours. I'll be recovering at the best grandmother ever's place for a while afterwards.

Stuff like TeΧ above isn't off-topic, this is open for personal ramblings after all, dear Benny. But I must say we prefer non-bzip2'd, i.e. gzip(1)ped or compress(1)ed, distfiles (zip, lzh are even better, because bzip2 is dead slow and memory hungry), but for e.g. iceweasel, it's okay to use bzip2 distfiles because nobody's going to compile it, at least not on a machine below 2 GHz, anyway.

If I can concentrate at all, I've got a few projects I could work at during the next days. One of these is an ncursesw mirport (including a "wide" flavour for the lynx mirport), another would be to re-work the, at least i386, bootloader interna (see below). Also, I've got an idea, sorry Benny, how to do a new website for us, which I could code first, propose then. Maybe I'll hack a bit on it. It uses a totally different approach than both www and web modules.

Bootloader… yeah. A few ideas: make stuff like PBR/bootxx, and the two MBRs, shell scripts that install themselves, nuke installboot. We need a C language helper for looking up block numbers (for ffs) but e.g. CD-ROM or raw disc image installs can do without. I'll try to not add MirOS specifics, so it could be run on e.g. GNU/Linux too. Also, I want a micro-PBR which just loads another PBR from say, sector 15 of a CD-ROM, which fits into the first 128 bytes of the first sector, so it can coexist with a Sun disklabel. Then, the bootloader could determine by reading a sector from the disc whether it's El Torito or not, so we won't need mkisofs(8)' -boot-info-table option any more and can use an unified bootloader for Live-CD and Live-HDD ;-) boots. That one could, then, also make use of MBR partitions – for example so we have a "boot hd0a:/bsd" as usual, "boot hd0,3:/bsd" for MBR partition (that's the only way to boot off a FAT partition if the disklabel doesn't have a slice for that filesystem), "boot hd0,2,0:/bsd" for extended ones. A "machine partinfo" command to display MBR and disklabel entries would, of course, make sense and complete that all. Finally, if I get the new installboot(8) replacement to be endian-independent and to convert the sparc one as well, dual-boot CDs will be much easier. (The sparc stuff needs a fake disklabel, but we could even hardcode that, much like the hardcoded MBR for macppc.)

While I was in the train today, I looked at some of the more arcane inner workings of MirPorts: internal dependency handling.

Packages have dependencies. There are three types of them in MirPorts: build, run and library dependencies. I won't talk about the former here. MirPorts inherited a very convoluted way of handling those dependencies from OpenBSD, including several passes over the packing list using various scripts and the infamous pkg utility. pkg is essentially undocumented and constitutes a failed attempt of Marc Espie to rewrite the pkgtools in perl. It has only one subcommand, pkg dependencies with three sub-subcommands show, solve and check.

Dependencies are declared via the LIB_DEPENDS and RUN_DEPENDS variables in the port Makefile, for example:

LIB_DEPENDS=           apr::devel/apr/apr

From there, they are passed to a recursive solver in bsd.port.mk, which outputs commands like

@libdepend apr-util-0.9.12-0-bsddb:apr.0.9:apr-*:apr-0.9.12-0

into w-*/pkg/depends. The four arguments are called name, libspec, pattern, and def. The @newdepend directive is for run depends and does not have a libspec. The depends file is fed to pkg dependencies solve, which replaces these lines by

@comment libdepend apr.0.9:subversion-1.4.0-0-bsddb:apr-*:apr-0.9.12-0

Note the changed order of the arguments! What it also does is resolve the dependency and write a static @pkgdep line like

@pkgdep apr-0.9.12-0

on top. All of those are output into w-*/pkg/PLIST. Upon installation, pkg_add only cares about the @pkgdep directives.

There are a number of problems with this approach. I would like to replace pkg entirely by C code inside the pkgtools. The advantage is a much reduced complexity, the removal of the liability that is pkg, and more flexibility: If packages are upgradeable, the final resolution of the dependencies should be done by pkg_add when installing the package. If you have a newer version of a dependency installed, we can check if the library version matches and still allow the installation. pkg_upgrade would also profit massively, as the current implementation leaves the @pkgdep lines intact so that they are effectively unfulfilled. This project is promising but it will take me some time. Be patient.

If you want to discuss this, please use the miros-discuss mailing list.

I haven't written anything here for quite a while, so here is a little update. I moved out of France and now live in Saarbrücken. I will hopefully have both telephone and DSL service soon—it is scheduled for tuesday, but let's say I will believe it when I see it. I started work on a port of Qt 4.2.0, as close as possible to the upstream sources. Those crazy patches from OpenBSD often break more than they fix. Qt is quite a beast though: a 36 MiB .tar.gz with no .tar.bz2 available, non-standard configure, .la files generated by C++ code, etc. Now I know why I like Gtk+ better :). Seriously, the tarball gained more than 4 MiB since 4.1.x. I knew that Mozilla is bloated (Firef*x is ~36 MiB compressed) but I was shocked to see Qt is just as big.

And now for something completely different: While writing the report for my summer internship, I came across a number of problems regarding pictures and their insertion into a LaTeX document. I will note my solutions here so that I don't forget what I did, even if this is a wee bit off-topic.

First problem: Inserting a plot done with gnuplot. Obviously, there are many ways to do this but here is what I do. My gnuplot file looks like this (example):

set encoding iso_8859_1
set term postscript color solid eps enhanced lw 2 "Helvetica" 20
set output "sucres2.eps"
set xlabel "Temp{\351}rature ({\260}C)"
set ylabel "Heat Flow (W/g)"
set key top right
plot [150:200] 	"bs06.4Kmin.txt" using 1:3 with lines title "exc{\350}s en chlorate"

The "set term" line is the most important here. color solid gets me a coloured plot without dashed lines. eps generates a correct bounding box and makes the plot a nice 12x7 cm. enhanced allows me to write stuff like "CO_2" and accented characters: {\260} for example inserts the latin-1 character with the octal code 260, a degrees sign. lw 2 makes the lines a bit thicker for better visibility. The 20-point font is good if you scale down the plot (like me).

Now you'll need to transform this into a pdf file. Use epstopdf:

$ epstopdf -d sucres2.eps
* Input filename: sucres2.eps
* BoundingBox comment: %%BoundingBox:
* Output filename: sucres2.pdf
* Ghostscript command: "C:\Program Files\gs\gs8.54\bin\gswin32c.exe"
* Compression: on
* Ghostscript pipe: "C:\Program Files\gs\gs8.54\bin\gswin32c.exe" -q -sDEVICE=pdfwrite  -sOutputFile=sucres2.pdf - -c quit
* Scanning header for BoundingBox
* Old BoundingBox: 50 50 410 302
* New BoundingBox: 0 0 360 252
* Offset: -50 -50
* Ready.

Note the correct bounding box. For inserting this, I do:

\begin{figure}
\centering\includegraphics[width=8cm]{Plots/sucres2}
\caption{foo}%
\label{fig:sucres}
\end{figure}

Figures you have only as Word drawings are a whole different beast. But there is a way to do it: First, mark all elements of the drawing, copy it and insert it into a new document by itself. Important: You have to position it as close as possible to the lower left corner of the page, as this the origin in PostScript. Next, use File > Print, select a PostScript printer and check Print to file. Open the resulting .prn file in GSview (I assume you are under Windows -- you use Word, right?). Select File > PS to EPS and uncheck Automatically calculate Bounding Box. Select the margins of your drawing with the mouse and save it under a file name ending in .eps.

Now comes the interesting part: You open the .eps file in a text editor. Among the first lines, you will see a BoundingBox DSC comment, like this:

%%BoundingBox: 11 11 363 209

Note these values. Search for "PageSize". You will find a line like:

<</PageSize [595 842] /ImagingBBox null>> setpagedevice

Replace the values in the brackets by the third and fourth value from the bounding box above. Save and quit. Again, use epstopdf like above. Open the pdf in Acrobat Reader. It should have only your graph without any additional white space around. For the insertion into your LaTeX document, use the same lines as above.

High-End: we now run past 2038 as well: in 2100 (32 bit), almost no- thing breaks, except a few library functions and ffs (UFSv1) timestam- ps, in 2200 (33 bit), network and shutdown don't work any more, but we at least don't crash.

Weird-End: sysctl(3) ddb.crash now can force a kernel panic.

Low-End: I'm helping Vutral (real name known) to install MirOS on an 8 MiB RAM 80486DX system. That's fun, but the serial console is a slow UMTS link, which is NOT fun.

French-End: Benny apparently is in Germany now and of course, thanks Deutsche Telekom, he doesn't have internet access. *sigh*

Data-End: I lost herc:/home and one of two underlying parts of raid0 (RAID level 1) alias herc:/ – and as if that would not be enough I had an rsync accident (missing directory level on destination combi- ned with --delete) losing most of the stuff in my $HOME which was in a directory on /… note to self: dont keep backups within $HOME but use a dedicated directory tree.

GNU End: gcc miscompiles not only C++ but also C code with a certain option -fomit-frame-pointer – caught by my favourite text editor.

Money End: At last, I get some state money (by far not enough), more might come, but I'd prefer a job. OTOH they'll pay for new teeth, more than they had paid if I were employed. But the pain's the same.

Advertising pays off, the Sourcemage guy (morfic) not only fixed the spell (their lingo for port), but also threw its name into some random IRC conversations in places I don't dwell. Cool.

I got a surprise when, preparing the new snapshot with Live CD to be issued, the BSDstats.org code didn't run any more. Luckily, we got the bugs and updates past that state. Getting ftp(1) to not pretend it can handle HTTP/1.1 responses did surely help.

Installing Windows® 2000 on a Pentium Ⅰ notebook without a CD-ROM drive can be entertaining. While it's a real PITA, having MirOS at hand eases the pain very much. (And I discovered a local DoS attack in our kernel, default installation, during the process.) However, the ancient game (Age of Empires Ⅰ) I was installing… I needed a no-cd crack, and all failed but one, which was merely a text file of 15k credits and four "patch these and be happy" instructions – what I did. Now, which editor can I use on DOS for a 1'622'016 bytes executable? Volkov Commander and debug.exe fail, so I used jupp. Yes, right. A Cygwin compiled (4 files: libcygwin, libncurses, jupprc, jupp) executable of my favourite text editor. And it worked for me and saved my arse from the pain of copying back and forth, without a putty installed (yet – will follow). To only change a 7504 conditional jump into an EB63 at D769F and adjust F7D8 at D7706 to 4090 is enough, contrary to the beliefs of people making 500+ KiB RAR/ACE files with a PE binary (can you spell trojan?) inside. Anyway, the laptop initially wasn't intended for gaming but for office work under Windows. However, the hardware (Compaq Armada 4220T) is cool, so I saved some 384 MiB on wd0 for a full (yet kept quite minimal) MirOS+X11 install, using a DOS PAGEFILE.SYS for swap.

I spotted quite some bugs working on the snapshot – I believe, it's already in a much better shape than #9. And I kept a few great ideas for past the snapshot, even. (So that people can upgrade a running system to get the security fixes, some of which I incorporated before That Ơther BSD.) What I didn't spot is Benny.

Did some serious FreeWRT hacking at wbx@' place yesterday. He showed me how to roll my own Sushi and we had enough wine to get me a serious hangover today.

Benny and I agreed on renaming #9bis into #10 because some seriously deep changes which could not entirely be avoided. This way we can also polish a little more and make one or two user-visible changes, but I'm trying to not delay the release any more for that reason. Afterwards a #11 will be released, with the goal "sync with OpenBSD 4". I also have plans for #12 which will take quite a long time.

Someone remind me to get Darren Tucker a beer.

I need plumbing in Plan 9! If I select a line like

	/usr/src/foo/bar/bar.c:123: warning: fnord

then I mark the filename with the mouse, type "jupp ", hit the middle mouse button, press return. That could be easier. Plan 9.

A person asked me in private mail how he could use the root account, on his freshly installed (in MS Virtual PC) MirOS. I told him that one uses sudo(8), but I still wonder why people do not ask in public, on a mailing list. Even if you ask in German there, you won't be considered impolite because about half of our usership could still make use of an answer in that language. (Of course it's better to write English.) But keeping these mailing lists and the archives is work, so, please, if a question appears, use these, or at least IRC.

Update: looks like it's greylisting. If the 451 is received the user gets a failure reply even though the message will be sent correctly ~4 hours later. If everyone would just be using IPv6… at this rate, spammers will be using IPv6 before the rest of the world. Even the uni of Düsseldorf.

make(1) needs a break-and-continue mode. Oh, and, if you change some types in the compiler, make sure to re-build and install all libraries before rebuilding the system, otherwise xlock might not link because a symbol was missing in libstdc++ (do I need to change the shlib version now? no idea, I think nm might know).

I hope the last wlog entry did not scare too many of you. First, I'm apologising to Benny for slandering his effort. Second, MirOS is going to continue, and in fact, Benny and I have talked about long-term plan (not 9) regarding packages, syspkgs, and so on. This is, seriously, an idea that will take years to complete, so we won't discuss it for now, unless someone asks us (over a few beers *hint hint*).

While my work situation did not improve yet, I hope to be invited to a personal talk at some companies soon… who knows…

I hacked quite well on fwcf – our first hosted foreign project – that's some good creative work, finally.

mksh(1) world domination is going to succeed.

-rHEAD contains lots of security updates; yes I promised a snapshot, quite some time ago actually, to be out shortly. Okay okay.

The more you work with software, the more it sucks. I found out that I have to rewrite at least mksh(1)'s emacs editing mode, due to design bugs rendering command-line editing with tabs broken, but better re-do the whole damned thing (taking 64-bit integers and wide chars into the initial design considerations). Same goes for MirPorts, after stumping across a bazillion more quoting bugs. If I had the money…

À propos rewriting… don't you dare mention FreeWRT.

A computer-illiterate visitor today decided the new design sucks too and prefers the old one. Yay! I still wonder why we didn't keep my CMS (or rather, excuse for not having a CMS) and changed only the content; dynamic page elements, such as news or wlog, could've been implemented within the existing framework as well.

Yep, MirOS now hosts foreign projects within its CVS. Trusted people only, of course. (Time to redesign CVS's operation.)

I'm missing a lot more catch-up things but you should have a look at BSDstats and their sub-page over MirOS.

Somehow I've been missing "the spirit" lately. Add to that the truth that, out of MirOS, I've basically written nothing, only doing integration work or writing (sometimes larger, such as with e.g. djbdns) patches. Even these rarely add major functionality. Magic is only contained in the one thing I have really written myself and of which I can be proud, the licence. Sucks, if you consider yourself not a lawyer but a programmer, eh? On the other hand, integration work is, both, importand and a good starting point into snapping how all works. Maybe I should really "clone" MirBSD again, this time not as an OpenBSD fork, but as a modular smallish system not even self-hosted in the beginning. Starting with the cvs- and ctm-based infrastructure, then some kind of packaging tools, initial utilities including an mksh rewrite, a bootloader, kernel, installer, third-party apps. But that's a thing I can't do alone unless I dedicate myself fully to it and some other person maintains MirOS. I guess I will never get that far. Wise, it is not, to reinvent the wheel. Building upon and improving existing solutions (which suck) is the way to go in this world. Does this suck? I guess it does. And I guess I'm not the first dreamer.

And I guess I don't have the vigour to do even one of them :'(

Yeah, to hell with OSI, the FSF now has two webpages which are under the MirOS licence: mksh's page and mirmake's page, and the page of jupp the editor. Cool!

We have a new design, I must say I'm… shocked. (And make these <h2> smaller in the CSS, please.)

Other than that, not much noteworthy happened.

Okay, the new website is online, with fixes. I don't think using other peoples' design is a good thing, because, if you don't use a design of your own which you don't licence to others freely, you aren't safe from copycats. This is a reason for the "must not copy the design of MirOS or (my) personal webpage" clause in the old www/LICENCE file, by the way. (I don't like the news entry format much – it has too much red tape (everything before, and especially the colon-space (use an equals sign, if you're going to use a key-value pair at all!), in the header).

Played with MacOS 7.5.3 in Basilisk Ⅱ and SheepShaver today. I think the GUI is quite intuitive. The OSnews article on GEOS was quite the good read for the evening too, and fasm (flat assembler) looks not bad either, and it has a funny BSD-style licence.

On the other hand I ended up doing my own letter form for a DIN 5008 letter all afternoon, because there are no good ones available, and no single one for Lotus Word Pro 9.5 (Millennium Edition) for Windows, at all! This sucks, the freely available specs (the standard itself costs some 30-50 €) suck and contradict themselves, but IMHO, I crafted something usable which doesn't make me look bad to my hopefully future employer at the first look. (On related news, I am still looking for a job, anyone?)
PS: Long live (AFPL) Ghostscript!

I'm pondering to release mksh R28 now, codename Pleitegeier (not an austrian word though). I should move some of the MI Plan 9 support code into the branch though, as I'm suspicious of it causing issues with job control in Plan 9 itself. The FSF and UNESCO directory of Free Software sadly still doesn't list it.

I should convert the entire repository to strict ISO_646.irv:1991 or UTF-8 conformance. Maybe post-#9bis, because I'd really like to get it out of the doors, but then really (including change logs, mailx(1) and submit.cf default, etc.) and correctly.

Cygwin, Interix, various Mac and Windows instances, GNU and BSD suck all, just in different flavours and strengths. And Solaris is the funny dot on the i. (What is Plan 9 APE then? I guess I don't want to even know.)

I started working on the all-new MirOS website in the all-new web module. At the moment, no new content nor design has been incorporated. However, I started writing a parser for news and wlog entries. These entries can be dynamically converted to HTML and RSS formats.

The entry format is largely self-explanatory and like an e-mail message. The header has metadata of the form "Name: value" and is delimited by an empty line. Then comes the body, delimited by a line with four hyphens. New entries are added at the bottom but I show an example of sorting antichronologically on index.php.

There won't be any more base64-coded filenames. I rather want to use php like you would use server-side includes, i.e. <?php require "header.php" ?>.

tyler@ pointed me to oswd.org, a website with free design templates. I have some ideas on how I would like it to look but nothing definitive yet.

Tonight I must write down my thoughts again, as to sort them and may bsiegert@ help me with some, as it touches MirPorts and the Macintosh. First off I found that ranlib(1) is needed on Darwin and added it back to <bsd.lib.mk> in an if clause. Now, we're already building the standard mirmake library in ${LOCALBASE}/share/mmake/libmirmake.a (the default location is different, but I'm focusing on MirPorts, willingly forgetting MirMake works on GNU/Linux too for now). I intend on adding a lot of functions to libmirmake – already added arc4random_push for systems which don't have it (i.e. everything other than MirOS), an arc4random(3) suite for systems which don't have that (Darwin didn't have it in older versions (I now scan with autoconf-style tests instead of using hardcoded #ifdefs) and the one of Interix sucks, and, it can be replaced by one spawning an EGD later), libohash, fgetln(3), i.e. everything which is required for MirMake anyway and more. Now, an enhanced libmirmake is not a problem. but not a benefit either, unless it gets used, which it doesn't, yet. Adding "-L${.SYSMK} -lmirmake" to LDFLAGS is a way to easily do that, but then we're entering the libgcc hell: what if we link libfoo, update libmirmake, and link progbar with libfoo? Suddenly, we have two different versions of libmirmake's code. Clash. That's why we need libgcc_s.so and libmirmake.dylib. Now that's not the problem either (I started versioning it already), but where to place the library? Can we use rpaths etc. so that on all systems, when we link against libmirmake, its location is added to the binary built, or library, of course, so that it can be found even if it's outside of any normal shared library paths and LD_LIBRARY_PATH variables? Or must I go the way to place it in ${LOCALBASE}/lib like any other shlibs? In the latter case, we could probably do a two-stage build of everything, leaving the Interix bootstrap (NOMAN) case up to the caller; that's an interesting yet slower (in compile times) outcome. Or did I overlook a totally different thing?

Maybe we should call libmirmake libshouldbeinlibc – oh wait, a library of that name already exists in GNU OS. (On that matter, Debian currently investigates the case of libast, which exists twice as well, once as "assorted spiffy things" and once in e.g. ksh93 (the shell), where the latter definitively has age precedence. I wonder how MirPorts is going to handle this and who's going to do it. Fact is, we can't have two libraries with the same name, and the devel/libast port is probably going to have to change its name.

While tinkering with the mirmake build system, I also added autoconf style scans to mksh's. And mirmake's build should definitively stop if errors are encountered. PS: Solaris /bin/sh is… weird.

Developing on Windows® 2000 is different. Not that productive if you ask me, you're much more limited if you are used to terminal-based operations. At least, the combination of PuTTY, Cygwin (ssh, scp, cvs, working text editor whose poll loop doesn't, like 16-bit DOS jupp, get my laptop to turn itself off due to heat), and Interix provides a work environment which is slightly suitable, but I still don't have perl5.8 for the mksh testsuite or stuff like that, and my fan is spinning. The hardware support of MirOS is so much better.

bsiegert@ might also want to convert ports/Hints into something more usable to the end user, and we had a request for MirPort howtos on the mailing list today too. Sysjail 1 was released today, and I like the benchmarks, which show MirOS easily outperforming OpenBSD and even NetBSD® in most things.

Angelo (the MacWorm hacker) is currently in town. I got quite drunk, last night, I must admit. I had an interesting hangover today. Hacking helped me to overcome it though, as did much sleep (too much). This is not improving my bank account's or my work situation though.

During a Netzladen session @LUUSA, mksh(1) now works on Plan 9.

I've been hacking random stuff, improving and adding user feedback I got, but I'm still primarily looking for a job in the area programming or system/network administration. Contact me if you have any hints, in this region.

While we're tracking down a few remaining bugs in the package tools, our lynx port is the best one available for the Mac and I'm submitting offsprings (e.g. mksh, mirmake) to the FSF directory.

bsiegert@ and tyler@ started hacking on the new website.

I unbroke HEAD, so snapshots can be built. I couldn't yet finally do the required work on the baselive CD, due to distraction (chess with a friend who needed attention) in the evening. Expect a new snapshot (of i386 only) soon (sparc will take fairly longer).

tyler@ yelled at me in Jabber today to "STOP COMMITTING SO MUCH :p", and then begged for MirPorts ${SYSCONFDIR}/screenrc… seems as if he found a unique selling point, finally…

mksh is now probably ready to release R28.

If the person responsible for the existence of /usr/bin/libtool on a Mac, which is not GNU libtool, and that this pathname is hard coded in gcc's specs file, gets to meet me, in a park at dark night, I can't be held responsible for the outcome…

I hope src/distrib/ is in usable shape again. I'll fix missing parts and build a snapshot afterwards, maybe without live cd for now, though that'd suck. The sparc live cd is not completed, although the miniroot should be buildable now, for the first time in MirOS ever.

Despite it all, I made lynx and GNU groff mirports, reinstated tyler as developer by his own request, fixed *roff macros (new MirNroff cpio distfile for Interix might be required soon), enhanced and fixed docs, helped FreeWRT people (website/server is down, we know), etc.

Licence is now up to date again, and links are actually links.

Sendmail 8.13.8 is out, but not a high-priority update.

Today it was rainy, depressive, and headaches arrived. Not the best, but a chance to hack something. I couldn't do much, but I improved the shell again and got an old NetBSD® PR (30842) finally fixed.

I also cleaned up some of the stuff in src/etc/ (especially rc which should be faster and more reliable now). Works for me.

I've also started on reworking the text of the mksh page, but can't get myself to finish it. Call it summer hole.

Although I'm still looking for employment I can't get myself around, sending out applications etc., because of these bad circumstances like the weather etc. If someone wants to pay 1000 € netto (i.e. taxes and so on already subtracted), or more, for a unix developer or admin, please contact me.

I just came back from a prolonged weekend at my best friend's place. That's why there hasn't been much activity, except Benny's starting an update on documentation.

Hubert should just always use cvs -Rq up -PAd like every CVS user in the world (according to MirOS docs). I did think of implementing a tag in the output for sticky files for him, but I don't think it's worth a try because stickiness can actually be useful (tracking branches).

I did some major cool mksh(1) work. Please test the beta. Details in #mksh (freenode).

I'm off for the weekend.

Benny has started fixing the website, finally (thanks!)

He also committed a www/firesomething security update.

Yay for bsiegert!

I've played around and fixed a little too, not too much.

More fun with FreeWRT (it sucks) and IRC (we now bridge a channel in two networks, but the TCL plugin for eggdrop sucks worse, so ciruZ has threatened to rewrite it (including the bot) in Ruby – so gecko2 can't configure it to death and it won't talk).

It's still fairly warm.

herc.mirbsd.org now serves dynamic IPv4… TTL is 60 so beware, and it can't be updated in case herc goes down.

I planned to write a more lengthy wlog entry but since I have gecko2 here visiting me for a few days I don't get to do as much stuff as I'd like to (but on the other hand he helps me cleaning up the appartement but also makes me work :'( hard…). And now I even have headaches and get some VPN DNS whatever confusion instead of working on distrib. Also he tells me to goto bed at weird early hours, and so.

But the kitchen already looks more clean, and I did my job too.

It's too warm a summer.

I planned to write a more lengthy wlog entry but since I have gecko2 here visiting me for a few days I don't get to do as much stuff as I'd like to (but on the other hand he helps me cleaning up the appartement and spots a few rare MirPorts bugs). Thus, this must suffice:

I wonder why Benny is still using the ancient operating system, Plan 9, when there is an SSH v.2 client for the progressive operating system MS-DOS® – he should use that instead, so he could directly commit from there. (For the record, an MS-Shell 2.3, along with a few unix-like utilities, runs on 8088 CPUs quite fine, along with the best web browser ever (but if you have an 80386, you be better use the real thing, complete with HTTPS/SSL support instead). The site even has an entropy gathering dæmon (URANDOM$), so… Benny, now please get and install some DOS or my well-known image.

That MS-DOS shell above has got most (not all tho) mksh(1) features! Yay for the Korn Shell!

If you update MirPorts, don't forget to re-run Setup.sh(8)

I've finally got my lazy arse around to start the redesign and sparc enhancement of src/distrib/, but couldn't yet finish the work as ゲッコニくん wants me to sleep. I've even merged bsd.rd, floppy and iso9660 images generation into that and will try to make it even better. The sparc miniroot was, still, not touched by me though. Also, there's no live CD yet :)

Don't forget to always use irc.mirbsd.org:6668 to connect the Freeforge IRC network (use port 6666 if your IRC client does not speak TLS/SSL and you're too lazy to add an irc bridge in inetd.conf(5)). We still sit in #bsd and #mirbsd (as well as Freenode's #mksh).

mksh(1) is now /bin/sh, the other one is nbsh(1) (/bin/nbsh) for now until the bugs are fixed. When MirOS #9bis gets released, its codebase will be tagged as the next version of mksh (R27f) as well.

I probably forgot many thing but it's late and this FreeWRT thingy's been distracting me (let's let wbx handle stabilising ;) and it hasn't yet cooled down much either. So please forgive me.

Other than fixing the website ;) I did only some minor stuff just to keep activity. I also "have to" hack on FreeWRT but it's way too hot for anything productive.

GNU make sucks.

Benny should just use the US keyboard layout anywhere.

I had a hard time fixing FreeWRT, but I eventually succeeded. MirOS, on the other hand, didn't get short either: I renamed /bin/sh to nbsh, to have the option of changing it to mksh until the bugs are found and fixed (nbsh(1) will be kept anyway, so people can check their scripts, to make sure they still work with ash) while they're reported upstream (to TNF), and we got a pciutils (lspci) port. gecko2 installed MirOS 9 and current ports, and we thusly found a small buglet in MirPorts (GNU stuff) which got fixed as well. (growfs(8) works well.) I also did few more minor things to clean up and prepare for other changes.

It has been a rather long time since I have written something here. In the meantime, I moved to Saint-Louis and begun my summer internship. I am working on explosives, and dangerous ones at that. But the people at the institute are very nice and let me work freely on my subject. There is not much to do in the evening, as the town is small. Even if Switzerland and Basel are only a few hundred metres away (really!), I don't go there too often.

After all the interesting talk with Uriel on FrOSCon, I really installed Plan 9 on two machines and tried to work with it. In fact, I am writing this in the acme editor on Plan 9. I managed to get my own user environment as well as TeX running. There is even a web browser called Abaco, which does about as much as dillo but has only 5000 lines of code. It even does tabbed windows, acme-style. The only downside is that I cannot commit from Plan 9 as the ssh client only does ssh1.

But even with all the other stuff, I still do MirOS and MirPorts development. I am on a modem line here so don't expect stuff that requires huge downloads. I have been working on wxGTK 2.6.3 with the libtool patch from pkgsrc. So far, I got segfaults in gcc and even in /bin/sh during the build. I don't know why but the Makefiles expect $(top_builddir) to end with a slash, which terribly confuses the poor shell. I also have distfiles for cairo 1.2 and gtk+ 2.10 on my machine, so this might come in the not-too-distant future. I finally put something into the aqua category: AquaTerm. However, there are no other ports that use it at the moment. math/gnuplot will pick it up if it is installed, I am changing that.

I am getting slower with all the typing because I am constantly changing keyboards between German, French and US—the latter for example here in Plan 9 because I don't know how to change the keyboard layout ...

I started to hack on MirOS again – despite the heat – on the distrib stuff, specifically (using makefs(8), and synching between the architectures), wbx@ called me out for some FreeWRT work *sigh* (I have some TODO there too). So that's it for today I guess. But I'll be back.

Benny finally inaugurated the aqua category in MirPorts.

Hm… hacking on FreeWRT… bad code quality, although I may not rant as much as I want because, as a core developer over there too (yes, I really am one) I represent the project *sigh* but at least the rants won't be found here that easily.

paxmirabilis (MirCpio, aka pax(1) cpio(1) tar(1)) is now being used, instead of GNU tar. It got some enhancements (eg. extracting to stderr or rough glibc compatibility) in the progress and might include direct gzip I/O, and even ZIP (maybe zip, at first only unzip) support later. Yay for MirOS!

My idea is to replace busybox' cpio/tar/pax by paxmirabilis as well, but that will probably only be an option. No idea which one is smaller though.

We also plan to make the packaging system more alike to MirPorts, to reduce overhead and improve legibility. Targets such as update-patches will be added, and the other targets renamed too. In general, we would like to clean up the mess left there…

Maybe it could generate Config.in and the ipkg control file from the Makefile then (for the developer to check in, though) kind of like the DESCRIPTION and CATEGORY fields in a MirPort.

There's much more stuff to do, but for now I'm cursing about da shit (Linux® / iptables) being unextractable on a case-insensitive fuck filesystem like the damned Macintosh have.

Ew, yeah. ncurses was being configured for the build system, instead of the target system. Many ports call autoconf (seen by the message to please provide AUTOCONF_VERSION environment variable), or are in other respects plainly broken, or try to access /dev/pts or something. A few ports still use host tools (e.g. openssl uses nm).

For MirOS, I should plan to feed back binutils and gcc-4.3 ports (to cross-build from lunox to MirOS, at least) to the FSF. Maybe we could, then, even make a Soekris FreeWRT with MirOS kernel and BSD libc? That all isn't even limited to WRT boxen anyway, we plan much more.

Oh yes, and the entropy dæmon… need to investigate about how *nix IPC works best though.

#include <rant/usual/subversion>

If I get the clues about when mksh segfaults, we should bring out an mksh R28 with the fix, I think. Optimally so that mksh R28 is the code which corresponds to MirOS #9bis. Other fixes, such as manual page, or so, are of course greatly appreciated. Send diffs ;)

Most funny thing of today: Mac OSX Tiger comes with /usr/bin/make as GNU make 3.80, but this apparently is not "new enough" for a certain package manager, namely Fink, which brings in its own version, GNU make 3.79.1 – do I need to say that both don't run correctly and the MirPorts Framework is required for GNU make 3.81 on Mac OSX as of now? Besides… someone tell the FSF that case-insensitive file systems are not only occuring on DOS/Windows®!

Thanks to Dr. Pfeffer (pepper) of BOSng, I've now also been out to a geocacheing session for the first time, with his Windows® PDA that can do GPS. It took us quite some time, but it was helluv fun and I hope we'll continue the journey.

(statpics moved here)

Sourcemage GNU/Linux users might now also be able to use mksh(1) and Debian GNU/Linux has accepted the update. After mksh R27e though, I've fixed the regression testsuite to pass if invoked as root, which a few people seem to require… these who never learn… furthermore I got rid of a few bytes in /etc/profile on FreeWRT (and thus in ports/shells/mksh/pkg/DESCR too), using the local keyword, ipv the typeset keyword.

We now have adduser(8) and openssl(1) asn1octetstream (which is also documented in openssltool(1), and will be fed upstream), to complement my vnconfig(8) -f addition. That was quite some hard work, done with a bottle of Met and bottles of beer shared among us at yofuh's place. It is HOT!

I also fixed contrib/samples/etc_profile for foreign hexdump(1) that doesn't support %_aX, added some acronymes, and benz did cups. The web page of mksh got pushed up to -current and a text about its requirements.

Adding the above functionality is a milestone, btw.

The story on Daily Dæmonnews I submitted today got accepted!

surrounder from #UnixNL helps seeding BT, dankjewel!

I did get rid of binstall (the joke) now. And also, fd0 on machines without floppy. boot.net is also superfluous, as the elf2aout(1)-generated boot can be used as netbootloader as well (and quite well) – why not, since bsd.rd.net is created just the same way.

Benny now has internet access at home. I still am trying to prod him to commit that all-japanese wlog entry of his, though. (Wait! That one would even be committed to /?wlog-8 right? Damn boy you're late.)

It actually is cooler than the days before, good.

mksh R27e is spreading slowly, but the goal of mksh's world domination is getting closer…

Originally I had decided to go to bed earlier today, but somehow I'm fascinated by the just-finishing (tarmk stage) sparc snapshot build. I will publish that snapshot on BT now and afterwards try to hack on the miniroot and live CD (text-only for now) so it won't matter if I break stuff. Then I'll put a bigger HDD into the sparc and reinstall.

Some of that old code is so scary…

makefs(8) from TNF has appeared, and I already have filed the bug toolchain/33924 in their gnats ;) I hope they'll help improving it so we might someday get rid of mkisofs(8) in base. I also opened another report – toolchain/33925 with the filesystem size off by a block issue from 2003 or so.

I haven't been able to work much on ports, as expected. Benny on the other hand is currently trying to buy internet access via a modem. Yes a (USB-attached soft)modem, that analog phone line thingie. He'll work a little on ports then, but don't expect too much from him either.

I might also have some real-life things to take care of. I still am, additionally, looking for a job. Offers and donations welcome.

IPv6 tunnel providers suck. All of them. And perl(1) sucks more.

It finally cooled down today with a thunderstorm. Problem: both soil and air are wetter now, so if the sun comes back, it'll be sticky.

It's been even hotter, thus I'm still not sleeping (I'm 7 hours past midnight already) but enjoying the only cool hours right now. Insects, these beasts! They've been plagueing me worse than ever tonight.

Yay! The libc major bump turned out to be necessary, an API changed, an API vanished. And the sparc snapshot due will even have both shared libraries – rebuilding everything takes too long and I so dearly want to get rid of NFS. At least, I'm now already diving in the depths of src/distrib/ and will try my best to make as few regressions as can (no miniroot yet, as time permits).

I wonder if we can use NetBSD®'s makefs(8) for the floppies (any case, it'd just rock!) and maybe mr.fs and miniroot9.fs too (with some flag to make space or something). One also should look at its ISO 9660 building capabilities later (for MirOS #10 then).

I guess I'll have to bring out an i386 snapshot too, after the login manager changes I suggested to miros-discuss@ the other day, and to be fulfilling the "debt" to bring all the security updates out. The overall code is still sort of flaky, but improving. I wonder if it would be better to first fix everything known on sparc (including that miniroot, and maybe even a text-only live CD) beforehand, as to unveil pending MI bugs.

Is anyone reading this here anyway?

I thought it would be a good day, fixing lots of sparc and MI stuff, hacking on mksh(1), hacking on myBB for a friend, hacking on FreeWRT's mksh (making it cleaner) and the profile, maybe Debian mksh update and add sample ~/.mkshrc – the sparc is doing fine – but SOL I am… the internet connection went away.

What do you think about the following

	tg@demo:~ $ cat x.pas ; gpc x.pas ; ./a.out ; file a.out ; ldd a.out ; dmesg | fgrep cpu ; uname -a
	program Hallo;
	begin
	  writeln('Hello, World!');
	end.
	Hello, World!
	a.out: ELF 32-bit MSB executable, SPARC, version 1, for MirOS BSD, dynamically linked (uses shared libs), not stripped
	a.out:
		Start    End      Type Open Ref GrpRef Name
		00000000 00000000 exe  1    0   0      a.out
		1725c000 17899000 rlib 0    1   0      /usr/lib/libgpc.so.2.1
		23735000 23d57000 rlib 0    1   0      /usr/lib/libm.so.5.0
		16b9b000 1725c000 rlib 0    1   0      /usr/lib/libc.so.38.4
		21750000 21750000 rtld 0    1   0      /usr/libexec/ld.so
	cpu0 at mainbus0: TMS390Z50 v0 or TMS390Z55 @ 75 MHz, on-chip FPU
	cpu0: physical 20K instruction (64 b/l), 16K data (32 b/l), 1024K external (32 b/l) cache enabled
	MirBSD demo.mirbsd.org 9 Kv#9s81-stable-20060627 GENERIC#92 sparc

(ciruZ thinks it is pretty cool ;)

Okay, there's not much more to do except letting the sparc compile a full system during the night and installing that, as well as rsynching the repo to my laptop. I hope I'll soon be at the point where I can do the HDD swapping and get rid of sucky NFS.

I have a mosquito lamp now – UV light and a high-voltage thing killing them. But they're flying away from that and around in the room as if it were disgusting them. *sigh*

After backporting and fixing stuff, this is mainly unbreaking -rHEAD and letting changes settle down. xterm didn't get updated (yet – maybe xterm-213 works better than xterm-215, but for now we still have xterm-200), half of OpenBSD's XF4 fixes were added and some of our own devised (especially since neither XFree86® 4.6 nor x.org trunk can be built with freetype 2.2…) and things simplified. Now, I build everything on i386 (in order to check for regressions) and sparc (want to add a 9 GB disc drive and install a snapshot on that, to get rid of NFS, but there's no snapshot yet). Also a.out debugging symbol handler is back in kernel (sounds like Japanese Engrish eh?) since I fixed the point of breakage.

MirOS boots just fine on a Thinkpad X41, just the Broadcom shit does not like to work. I may backport that, says Benny who just moved to an appartement in the Germany-France-Switzerland border region and except not having a phone (tomorrow) and internet (modem) is doing fine.

Reading more code for FreeWRT… it's cruel. configure is mixing build and host system up – look at libdnet, but keep a vomit bag handy.
By the way… many people dislike subversion, like me, especially after having used it for a few weeks.

Plans? What plans?

I could do an auto-livecd for sparc too. Easily. But mine is running on VT420 on serial console. Also, it is slow. So spare me. (I offer to still do it if users request it. Don't forget releases are no live CDs.

Very late edition, a.k.a. WTF is date(1)?
I'm through with i386 and committed the distrib/lists changes, where I also shuffled some of the things I know won't run on sparc away into md.i386 files. Some has been disabled, even though it builds on sparc, like iopctl(8), because the hardware just doesn't exist. I then screwed around a little in gcc and found out we're still building a threaded gcc even though that's a java only thing, and optimised Makefile.lang (not yet tested). Note, a threaded gcc probably doesn't work on sparc. I decided against keeping the libobjc_gc wrappers unless manually (NO_BOEHMGC=no) or via depends (NO_JAVA=no) enabled, and for keeping c++, Objective-C, and GNU Pascal available on sparc, because there's no reason (except size which is no issue with gcc bloat, er gcc 3, anyway) to differ between platforms at all. They're small enough to not slow down by more than a few hours or so of build time ;-)

My sparc has just gone NOPIC=no, i.e. builds ld.so(1) and shared lib stuff just fine. bsiegert@ isn't (sadly) reachable by phone though, so I continued to hack on autoconf/libtool, regenerated all the stuff and then threw in the new perl. Now I've got to, after refining a few more code pieces, check if everything (including X11) builds on i386 again. On sparc, I've got to take what I've got, build the distrib/ stuff and a snapshot in order to install that on another hard disc.

I think -rHEAD is in pretty good shape, I'll upgrade as much as I'll be able to do in the next days, try to get a stable sparc development, build and test environment, and fix everything outstanding in the base system. Regarding ports, I won't do much but leave that to Benny, I'll do base system in the meanwhile and hope he fixes these ports which do need it. (autoconf 2.60 is in, but I don't think there's any fallout.) Thus, we can (tentatively) release #9bis in two or three weeks.

It's family weekend for me, don't expect too many commits. Also, I'm in that "computers suck" mood again but the weather (too hot) keeps me from giving in to it.

On the other hand, the Netzladen (usual place) has got me back… at least for drinking some fine Met (honey wine).

Benny, when are you going to commit that full-japanese wlog entry of yours?

I've built MirOS (base and gcc) entirely on the sparc and have fixed the GNU Debugger in the progress as well. I'll have to change a little code in order to silent gcc, but that's not much left.

In order to fix a few security leaks, I updated libpng and freetype2 as well. I yet have to re-build the entire X-Windowing system in order to test whether this actually does not break anything. Neither has the XFree86 environment been ported to the new sparc system.

After the bump of autoconf to version 2.60 everything must be tested and libtool ought to be bumped as well.

Finally, I must make sure that distrib/ works on sparc and release a first snapshot – optimally as live-cd :) but I'm not going there yet.

On a rainy day, I'm continuing to hack on sparc again. I haven't yet found why dynamic linking breaks, just how to break it even more, plus without a working gdb there isn't much I can do. On the other hand, my kernel didn't panic yet, so I'm going to make it a static arch for now and get back to ld.so(1) support later.

On a related side note, we're currently developing #9-stable in HEAD and will branch that later. I want a #9bis to be out both as soon as I can and sparc works reliably, and with as few differences from #9semel as possible. This also gives us a chance to polish ports and docs.

Benny is moving homes at the moment.

Oh, and FrOSCon was a full success, best event I ever attended. I've intended to write more here but I'm still done with.

Which <censored/> set the maximum datasize for sparc to a mere 128 MiB, in contrast to i386's 1 GiB? Bumped to 320 MiB in order to be able to build gcc… kernel really seems stable now, compiled both cross and natively.

I'm committing this as 2006-06-25, but actually, I haven't slept for more than 31 (thirty-one) hours now, so it could very well be under 2006-06-24. I'm going to sleep now, but not without releasing an ISO of MirOS #9semel on BitTorrent.

There's now a new BitTorrent info page not linked anywhere except from the tracker (and here). It ought to describe what the torrent will give you.