Casper Dik's Weblog

The End of Realmode Boot I've already mentioned two great new features in our current development release; ACPICA and USB hotplug.

But there's one change that's much more far reaching than that: Newboot.

Most Solaris x86 users will be familiar with the blue screen/device configuration assistant/boot sequence and how ancient some of that feels. Perhaps few are aware that the DCA is actually a realmode DOS like environment where each boot device requires its own realmode driver. These drivers needed to be compiled with a 16 bit compiler and 16 bit MASM, not available for ready money anywhere. While the official build environment required NT, I managed to build it on environments ranging from MS Windows 98 and 2000 on actual PCs to Caldera DOS 7 on a SunPCi card (which allowed for automatic building which was great fun). Now that this piece of shameful history lies in the past, I am not afraid to confess.

But as of last Sunday, April 17th, 2005, we have "legacy free" newboot. Newboot uses grub with ufs support so we now have native grub support and a menu we can edit from inside Solaris. Device enumeration completely done using ACPI

Because it skip the device configuration assistant and boot a single large file with all kernel device drivers which makes startup quite a bit quicker and allows us to boot from any bootable device as long as we also support it in the kernel so we can mount root.

And we've reverted back to white on black consoles; this again takes some getting used, surprisingly enough.

One thing to note is that before you may had to disable ACPI in the kernel and the BIOS; with Newboot + ACPICA, you actually stand a much better chance of the system working with all the default settings: ACPI on, ACPI 2.0 enabled. Even legacy USB enabled now has a much better chance of working than before.

But this is a radical change an PC BIOSes and hardware being like it is, interesting times ahead. SO please test drive when this hits Solaris Express in a few months time.

As of this writing, it's a bit in the balance whether you'll get to see the source first as part of OpenSolaris or the binaries as part of a Solaris Express. (2005-04-27 00:00:00.0) Permalink Comments [14]

Netherlands/Benelux OpenSolaris/Solaris Usergroup A few of our customers approached me to start a Solaris user group in the Netherlands (or perhaps a somewhat larger area)

Any takers? Perhaps offers of venues, talks wanted?

For those who don't know, I am based in the Netherlands. (2005-04-26 13:38:40.0) Permalink Comments [5]

Yet Another Desktop/Laptop Usability Step "Solaris Nevada" build 14 is proving to be another quantum leap for Solaris desktop usability.

I discussed the new USB hotplug support in vold before, but in the last few days we've also gotten the virtual keyboard/mouse driver in the next Solaris release. People often complained about the fact that their laptop keyboard died until the next reboot when they plugged in a USB or other keyboard. Well, not anymore! We now have virtualized keyboard and mouse drivers which collect events from all available keyboards and present them through a single virtual keyboard and mouse. It is also still possible to use the devices as seprate devices in case you have a multi-head/multi-user environment, but for the common case of a single system with multiple keyboard (laptop + keyboard) this is another big step.

You can plug in the other keyboard at any time, running under X or the commandline, it just works. (2005-04-26 04:49:55.0) Permalink Comments [2]

Solaris FAQ Updated For the first time in many years (2.5 years) I've updated the Solaris FAQ

Much more work is needed on it but at least this is a start. I'm hoping to update it more regularly now. It's also still here but it seems to be doing fine there. (2005-04-26 04:45:12.0) Permalink Comments [3]

ACPICA in Solaris With the long history of neglect that Solaris on x86 endured, quite a few components got to be extremely stale and fragile. And this wasn't just a lack of device drivers but also a lack of basic new functionality in the core OS.

This week saw another quantum leap; the induction of Intel's ACPI reference implementation (ACPICA) into the next Solaris release.

For years I wanted to have battery support on my old VAIO and later on my Ferrari. And I wanted a power button that did something, etc. I tried to make do with the old "acpi_intp" interpreter which was part of Solaris; but it leaked memory like a sieve and was limited in functionality. Integrating ACPICA looked daunting but fortunately someone made an actual project out of this and the end result is that we now have a state of the art ACPI interpreter in Solaris.

There are basically only two ACPI interpreters in widespread use: the Windows one and the Intel one; by leveraging Intel's source, we stand a fair chance of having Solaris work with more ACPI BIOSes. If our system required ACPI to be turned of for Solaris to work, you may find yourself forced to switch it on when you upgrade later this year.

I've been distributing acpica and a number of other useful Solaris binaries in a single internal kit called "frkit" (originally aimed at Ferrari's but now running on countless systems); frkit includes acpica, a powerbutton/battery handler, an AMD PowerNOW! powermanagement module, a GNOME battery monitor, and our development cardbus and wireless drivers + tools.

One of the more interesting parts of that is possibly the "NDISulator" port from FreeBSD which allows the Sun Ferraristi to use the builtin Broadcom wireless on their ACer Ferraris in 32 and 64 bit mode.

ACPICA is just phase one of a larger project; we have not yet bothered much with the "P" (for power) from ACPI; but we hope to leverage the new implementation to provide the necessary "S3" and "S4" sleep state support.

The speed at with new features work on my Ferrari which I've had now for 4 months is in stark contrast with my Vaio which I got not too long before S9 for x86 was postponed. It's clear that we needed a ramp-up after the wind-down, but it seems to be going more quickly than ever before. (2005-04-20 02:46:56.0) Permalink Comments [10]

USB hotplug finally works Today I was pleasantly surprised to see the latest putback to SNV, our next Solaris release (soon on a OpenSOlaris source server near you)

Before this putback you could hotplug/eject devices into devices (SD cards and such in card readers, floppies in floppy drives) but you had to restart vold to mount USB pen drives. But now, you can insert them, remove them, etc, and they're mounted and unmounted automatically.

This was really one of my most wanted features and its great to finally have it. Will it make an update? I certainly hope so; but you can always try Sun Express. (2005-04-19 13:11:40.0) Permalink Comments [3]

Solaris 10 Encryption Supplement download The Solaris 10 encryption supplement is available for download here. Apparently, it's difficult to find going through the Sun download sides so I give a link here.

The supplement adds 256 bit AES and 448 bit Blowfish; DES, 3DES, 128 bit AES, blowfish and RC4 are already in the standard release. In other words, the standard release gets you what you would have gotten with the encryption supplement for older releases. And the S10 Encryption Supplement takes it one step further. The main reason for not making this part of the Solaris CDs is import restrictions, rather than export restrictions.
(2005-04-14 05:45:15.0) Permalink Comments [2]

Timezones and multi boot. One of the things that has always been bothering me is the fact that on x86 systems you cannot really run multiple operating systems and survice the timezone change. That's because the clock runs in local time; and localtime is ambiguous. The system cannot tell whether the DST change has been or not so it needs to record this fact in the filesystem (that's why Solaris on x86 has the "rtc -c" cronjob). If you boot all your OSes in turn after the changeover, your system will be N hours off once they're all done adjusting time. The problem is probably best summarized here

On Unix this was long solved by running the clock in the UTC or GMT timezone; that clock is unambiguous, give or take a leap second, and allows multiple versions of the OS to coexist.

Last week, it was pointed out to me in comp.unix.solaris that there is a hidden registry key in later releases of MS-Windows. I already knew how to fix up Solaris, so I combined this to:

       Set the following registry key (it does not exist!)

HKLM/SYSTEM/CurrentControlSet/Control/TimeZoneInformation/RealTimeIsUniversal
(REG_DWORD = 1)

       In the control panel with Day&Time settings, check the "automatically adjust" check box.


        Boot into Solaris and run:


              rtc -c -z UTC

        then correct the clock with date/rdate
        (if you use liveupgrade, lumount your other partition(s) and copy the
        /etc/rtc_config file to all of them)

        In Linux, you'll need to run "timeconfig" and select "RTC set to GMT".

Note that if you don't multiboot, it's probably also a good idea to run "rtc -c -z UTC" and then correct the date; for one you won't be bitten by the AMD64 timezone bug we had in Solaris 10. (2005-04-12 06:13:30.0) Permalink Comments [4]

Open Solaris CAB It has been a busy week flying to SFO and having our first CAB meeting. The first good thing that happened was that KLM had finally changed the aging and horrible MD11 for shiny new Boeing 777s with personal video. I have a bit of a problem with the new immigration procedures, and I like the Brazilian's government stand on this.
I had met Rich Teer fleetingly before in the hallways of the Menlo Park Campus so he recognized me when we checked in at the same time; we probably were on the same BART train from the airport. But I had never met the others. I feel we have a great team with very many different competences, from Roy Fielding's experience with Apache's governance model, Simon Phipps' tireless evangelism. And Al Hopper with his tireless Solaris on x86 enthusiasm. Rich Teer, a SPARC fan, and accomplished author and myself as Solaris engineering representative, being the more technical side of things.
Are we just marketing as the Register would have it? No, we're very serious about it. Is the CAB just a bunch of YES-men? Can we get the respect of our community if we are?
Sun takes both Open Solaris and the independence from Sun serious; Jonathan Schwartz came to meet us but none of the other executives was allowed at our meeting. He talked to us at length and was very serious about clearing up the roadblocks that we had already determined to be on the path to OpenSolaris. It is clear that they want us to succeed and want us to independent. Jonathan even stayed for lunch.
The Sun press conference we took part in was a first for me. The press was not hostile and mostly asked questions which were to the point; some more than others.
We have a lot of work to do and will do most of it in email on a publicly readable mailing list.
The second day we listened to Jonathan's keynote at the OSBC conference and spend the afternoon doing interviews with the press followed by a press reception and Sun engineering diner/Open Solaris launch party at Lulu's. And guess what, we were able to make the Americans walk all over town, the itenary was all "5 min cab ride" and some such nonsense.
On the final day I took Ben Rockwood's advice and tried out "Clam chowder in sourdough" after taking the cable car to lombard street and walking down to the harbor. The weather was gorgeous, the same cannot be said of the weather in Amsterdam which is now unseasonably cold.
(2005-04-09 05:44:32.0) Permalink Comments [3]