Older blog entries for spot (starting at number 104)

Aurora:

At this point, I'm left with the mouseconfig failure (I need to debug kudzu to find out whats wrong with it), the KDE audio failure (not showstopping for a beta), and KDE being slower than Debian Stable's release cycle.

Video works for my ss20 (GX), and my Ultra 2 (Creator 3d). Installing without /boot partitions works. Brian Ferris's keyboard "us" bug in text mode is fixed. All the devices are being probed and detected properly. All thats left to test is the merging of sparc32 & sparc64 on a single iso set. And I feel really good about that.

8bit GUI installs are ugly (not Aurora's fault, 8bit is inherently ugly), but they work.

Its good stuff. I'm really pleased.

Aurora:

Really fixed sparc32 (and probably sparc64) hardware probing this morning. Looks like it sees the videocard properly now too. Added some heuristics to tell anaconda to always try 24 bit video, since I'm stumbling all over sparcs that can't handle 16bit. It might be slow as hell, but at least it won't crash. (and 8bit is ugly, but possible)

It doesn't even pause in stage 1 now, except to tell me that its autoloading the scsi driver. Its so beautiful. :)

Aurora Bug List:

- mouseconfig seems to be failing noisily (but non-fatally) during post

- KDE is trying to use /dev/dsp, failing. Probably can be fixed with a symlink to /dev/audio.

- parted is complaining that the H/S/C counts on the sparc32 disks (older disks?) don't match what the disk label says... parted is wrong, needs patching.

- parted/silo seems to fail on machines without a /boot partition during install, although, the silo.conf generated seems valid

- KDE/QT is being poorly optimized by gcc 2.96, causing KDE to have terrible performance (5 min app load vs 30 sec for GNOME). This is not an issue with gcc 3.1.

- GUI installs always try 16 bit, and fail for majority of Sun video hw, added heuristics to anaconda code to try to make it smarter with this regard, for sparcs. Partially complete.

Aurora:

Did my first successful sparc32 install last night, on the dual SS20. Swapon still segfaults on sparc32 SMP kernels, but its not a showstopper. Parted still complains that the number of cylinders on the disk label don't match the geometry, but fdisk doesn't have any issue. I'm sure this is parted brokenness, because on the ss20, it was able to make/format partitions without issue. Videocard probing doesn't work at all on sparc32 right now, but on the good side, I'm pretty sure that I managed to fix SCSI & Network Card detection. Going to try to check that later this morning.

I'm might try to make another beta release sometime within the next week or so, likely just isos.

Aurora:

Work continues on sparc32 (I've focused almost entirely on it since my return from LinuxWorld).

Several stupid mistakes were present in the sparc32-BOOT kernel. Hopefully, I've got them all cleaned out now.

Also fixed the issue with serial installs not having ttyS0 in /etc/securetty (new setup RPM) and fixed the lingering tty bug in anaconda. Now, I'm just waiting for the kernel to finish rebuilding, and I'm going to do another sparc32 ISO spin. No, you can't have it now, its REALLY not ready for prime time.

Oh yes. I also re-enabled RAID5 in the BOOT kernels, and demodularized usb-ohci in the sparc64 BOOT kernel to see if it makes the keyboard magically work (since including the module in the initrd did NOT do the trick, as hoped). But I'm going to hold off on a sparc64 respin for a little while, so I can focus on sparc32. Maybe tomorrow evening.

Aurora:

"What? UltraSPARCs have SCSI?" ;)

Seriously though. I'm really really close to having ISOs I'm happy with for LinuxWorld. Jeremy helped point out the missing ioctl that was breaking the installs from yesterday, and I've been doing post-install cleanups all day today.

One big thing that was missing: the esp driver in the initrd. Thats useful for most UltraSPARCs, honest.

Box at home is definitely a dual Ultra 60. No cdrom in it though. Time to bring the external CDROM drive home. :)

I feel much better now that I've seen it install successfully on the Ultra 10.

Aurora:

*bangs head on wall* So close, and yet so far. The ISO boots (the biggest hurdle, really) kicks off the installer, gets all the way up to installing packages, and tracebacks. Thought the changeloop patch would resolve this issue, but apparently, it does not.

Acquired an Ultra 60 last night. The individual who gave it to me said it was a dual 30, but the case says its a 60. Haven't booted it up yet to check.

Starting to get concerned about whether or not the sparc64 cds will be ready by Linuxworld... :/

yup, that was me. NCSSM, class of '98.

I'm not terribly big on the bar scene, and while I enjoy a good game of pool, I dislike doing so in a smoke filled environment. That tends to cancel out a LOT of the options in Raleigh/Durham. If you know of a place that fits the bill, by all means, let me know. :)

I didn't even think about Sushi on the last post... Raleigh has some excellent Sushi restaurants (Sushi Blues, Waraji, Sushi Thai, Sushi Tsune).

*****

I'm holding off on Aurora posts for today, waiting for ISO respin.

Must see in Raleigh? There's nothing really in Raleigh worth seeing. I've lived in the Raleigh/Durham area for 6 years now, and besides some mildly amusing museums, a couple of half-decent restaurants, an IMAX theater, a mega-mall, and a few lakes, theres nothing really memorable.

There is also a little software company called Red Hat, but even that isn't terribly intriguing for most tourists.

Aurora:

Man. What a difference a month makes. In the course of a month, 20 bugs went to 10, then to 5, and now to 3.

One of the bugs is erratic (the installer freezes right as it tries to format, but it doesn't always do it). The other two (parted can't make a sun fs label) and (network cards don't autodetect in stage 1) are able to be worked around. All the graphics and help text has been redone, although, I'm sure the help text is... interesting in a few places. X configuration in the installer WORKS on my u10, which is shocking. And to top that off, text mode works too. I caught another bug in the 0.3 tree that I'm fixing now, and Jeremy suspects the erratic hard locks are kernel (VM) related, but it looks really really good. We're starting to talk about iso creation/testing good.

Good stuff. I just hope we can have it ready for LinuxWorld in 2 weeks.

To be clear, all of this is for sparc64. Sparc32 comes next.

And in slightly Aurora related news:

Congratulations go out to Ingo T. Storm, who is the proud father of a healthy baby boy. Don't stress over that HOWTO, we might not need it after all. ;)

Aurora:

If it wasn't for "work", I'd have lots of time to dedicate to Aurora stuff. ;)

I gave a very impromptu presentation on the origins of Aurora SPARC Linux, and the motivations that lead to its creation. I suspect strongly that I bored the audience to tears, but hey.

The good news? Two days ago, I did the first successful local "install" of Aurora, onto my U10. The installer didn't have silo in the base group in the comps file, so when it tried to run its silo magic to initialize the bootloader, anaconda tracebacked, but I was able to switch to the shell and install/run silo manually, and the system came up clean. Albeit, this is with the "base" set of packages only, and I haven't started to test X configuration through anaconda, but this is a huge huge step. Jeremy said he fixed the silo issues that we uncovered, so if I have time today, I'll try another install with the most current version of the auroraconda tree.

Mind you, this is still a ways off from "isos", so don't rush me with emails asking for that.

I've also been somewhat idly poking the 0.3 kernel with experimental optimizations, like the 0(1) scheduler port, rml's lowlatency patch port, and the super_page patch. I'm not sure I'll keep any of these things in the mainline Aurora kernel, but I might make a custom kernel available for people who are longing for such things.