Slashdot Mirror
Linus Moves To OSDL, Will Work On Kernel Full-Time
worldwideweber writes "With the announcement of the release of the 2.5.72 version of the Linux kernel came the news that Linus Torvalds will be leaving Transmeta for OSDL to work on the linux kernel full-time. The email calls this a leave of absence for about one year." Update: 06/17 17:19 GMT by T : As many readers have pointed out, the length of Linus' leave is not actually specified in this email.
From: Linus Torvalds
/how/ freely Transmeta has let me do Linux work. My email address will
;) press-release about this tomorrow morning, but I just
To: Kernel Mailing List
Subject: Linux v2.5.72 and a move to OSDL
Date: Mon, 16 Jun 2003 21:35:09 -0700 (PDT)
Ok, I waited too long for 2.5.71, so here's a more timely 2.5.72
release.
It's extra timely largely because the hash list poisoning found some
problems in the RPC code, making NFS break. Trond found and fixed the
breakage, so 2.5.72 should work fine in an NFS environment too. Let's
see if the list poisoning shows any other dodgy list users. Knock wood.
Also, Arnaldo has cleaned up a lot of the networking code to use the
generic hash lists, instead of the old ad-hoc net-specific list walking
code. That code has been tested pretty well, but please holler if you
see something.
Changelog for other details appended.
The other big news - well, for me personally, anyway - is that I've
decided to take a leave-of-absense after 6+ years at Transmeta to
actually work full-time on the kernel.
Transmeta has always been very good at letting me spend even an
inordinate amount of time on Linux, but as a result I've been feeling a
little guilty at just how little "real work" I got done lately. To fix
that, I'll instead be working at OSDL, finally actually doing Linux as
my main job.
[ I do not expect a huge amount of change as a result, testament to just
change to "torvalds@osdl.org" effective July 1st, but everybody is
trying to make the transfer as smooth as possible, so we'll make sure
that there will be sufficient address overlap etc to not cause any
problems ]
OSDL and Transmeta will have a joint official (read: "boring". You
should have seen the bio - that didn't make it - that I suggested for
myself for it
wanted to say thanks to Transmeta. It has been a special place to work
for, and hello to OSDL that I hope will be the same.
Snif. I'm actually all teary-eyed.
Linus
- tom -
Kudos to Transmeta for hiring Linus in the first place ( even if they did transport him to the USA in reach of overlitigious bastards such as The SCO Group ) and supporting his work on Linux for so many years.
OSDL is dedicated to enabling Linux and Linux-based applications for data center and carrier-class deployment. We provide the crucial hardware for testing and development at this level, giving open source developers around the world the resources needed to bring Linux further into telecommunications and the enterprise. We are an independently governed, non-profit organization supported by 21 industry leaders.
Sounds cool
I submitted this a few hours ago (always a bridesmaid, *snif*), along with two links not in the story above. One was to the NY Times story about it. The other was to this story which just came out at Wired . . .a brief interview with Linus about his efforts to stand apart from political issues surrounding Open Source, which refers to the
discussion here on Slashdot
about his opinions on incorporating DRM into the Linux kernel (among other things).
I'm just quessing here, but I think the following explains everything:
:)
'yenta' is probably just a some PCMCIA card driver module that you can load and unload. Unloading causes an 'oops' which means a kernel bug (access to some unallocated memoryspace etc.).
'jiffies' are one kind of time units. On 2.2 and 2.4 series, jiffies run at 100Hz - on 2.5/2.6 series, it will run at 1000Hz. It's just an incremental value that is used in time related jobs.
One related problem is a 'jiffie wraparound' that causes uptime to jump to zero again. This is only a problem on 32 bit machines of course. At 100Hz it happens somewhere after 400 days and on 1000Hz much sooner (in about a month IIRC). It's easy to calculate exact times if you happen to care
Kernel folks have also spent a lot of time lately to fix driver problems related to wraparound issues.
Sorry, should have added a link rightaway, so here it is. I think Linus didn't enjoy the 2.4 series as much as development kernels, and maybe also understood that others could be better at the more boring side of evaluating bug fixes etc. (Hats off for that!) He might share the maintenance with Andrew Morton, or possibly completely hand it over by 2.6.0.
2.5 is the development branch, it doesnt need "maintaining".
Well, it's already mostly in maintenance status, i.e. waiting for bugfixes, more testers reporting and so on.
An entry on a linked list is to be freed.
Upon it being removed from the list, it's pointers (prev/next) are 'poisoned'
by being set to hopefully really, really invalid addresses.
This is to catch any bad code that continues to use the now invalidated pointers.
The bad addresses should cause an opps instead of allowing the code to possibly trash other valid data structures.
Clarifications welcome.
You are being MICROattacked, from various angles, in a SOFT manner.
Here's the press release alluded to in Linus' email. Still no mention of his leave being limited to one year.
Linux Creator Linus Torvalds Joins OSDL
First OSDL Fellow Will Devote Himself Exclusively to Linux Development
BEAVERTON, OR, and SANTA CLARA, CA, June 17, 2003 â" OSDL, a non-profit, global consortium of leading technology companies dedicated to accelerating the adoption of Linux, and Transmeta Corporation (Nasdaq: TMTA), the leader in efficient computing, today announced that Linus Torvalds, the creator of Linux, will join OSDL as the first OSDL Fellow.
As an OSDL fellow, Linus will work exclusively on leading the development of Linux, the open source software that he created in 1991 as a university student in Finland. Torvalds will dedicate himself now full-time to guiding a distributed team of thousands of Linux developers around the world. At OSDL, he will have hands-on access to its state-of-the-art computing resources and test facility. He will also help set priorities and direction for the Lab's different industry initiatives.
"It feels a bit strange to finally officially work on what I've been doing for the last twelve years, but with the upcoming 2.6.x release it makes sense to be able to concentrate fully on Linux," Torvalds said. "OSDL is the perfect setting for vendor-independent and neutral Linux development."
Founded in 2000, OSDL has data centers in Portland, Oregon and Yokohama, Japan used by Linux developers around the world. With investment backing from Computer Associates, Fujitsu, Hitachi, HP, IBM, Intel, NEC and others, the lab sponsors key industry projects, including industry initiatives to enhance Linux for use in corporate data centers (Data Center Linux) and in telecommunications networks (Carrier Grade Linux). OSDL is increasingly being recognized as the center-of-gravity for the Linux industry: an important and independent central body that invests in the growth and innovation of Linux for the benefit of customers.
"OSDL is a leading Linux-industry advocate with the single-minded focus of accelerating its use throughout the enterprise," said Stuart Cohen, OSDL CEO. "Linus' decision to join us is a confirmation of the importance of our mission. OSDL is the only organization where Linux developers, customers and vendors can all participate as equals. The addition of Linus' perspective and guidance to the Lab will enhance our value to all three of these groups."
Linux is the fastest-growing operating system in the world. Revenue for Linux-based servers grew 62% in 2002, while overall sales of servers dropped 8%, according to Gartner Dataquest, a market research company. By 2007, Gartner predicts that Linux may grab 15% of the worldwide market.
"Linus Torvalds adds tremendous credibility to OSDL's efforts to drive the evolution of Linux forward into enterprise computing and carrier environments," said George Weiss, vice president and research director for the research firm Gartner. "The computing market is still questioning how far and how fast Linux can go as an enterprise-ready platform. With Linus at OSDL, many will be looking for leadership from the lab for answers to those questions."
Torvalds will join OSDL on leave from Transmeta Corporation, where he is currently a Transmeta Fellow. Transmeta is an OSDL member and worked with OSDL on the transition. "Linus has made substantial technological contributions as a member of our development team here at Transmeta," said Matthew R. Perry, president and CEO, Transmeta Corporation. "Transmeta appreciates and fully supports Linus' strong interest in devoting his attention and energy to certain emerging industry-wide Open Source initiatives at OSDL."
About OSDL
Founded in 2000, OSDL is dedicated to accelerating the growth and adoption of Linux in the enterprise. Supported by a global consortium of IT industry leaders, OSDL provides state-of the-art computi
Unfortunately, the CPU isn't the biggest power hog in a notebook, and their cost/power/speed ratio wasn't much better than slowed down pentiums.
I'm actually very excited by their technology. But the only Crusoe laptops I've seen for sale have had tiny screens and huge price tags. It would be less expensive to buy an iBook/PowerBook and virtual PC than most Crusoe laptops.
Also, why do you think Andrew Morton has started maintaining the "2.6 must-fix list" and had moderated two IRC conferences to know who will feel responsible for which lacking bug fixes etc.?
The piece mentioning Barry Diller, Howard Schmidt, and Linus can be found online here. It's the intorduction to their letters to the editior section.
Uh, where have you been?
Yes, OSDL did have Caldera/SCO money. Caldera joined OSDL early on, and later let their membership expire sometime after the name change to SCO. So SCO has funded Linux kernel development projects through OSDL. Larry
You are making a good point, but I think I can name three:
[ Note: I am only comparing the MS Product WinXP or Windows Server 2003. If that is to restrictive I imagine you'll correct me. Also I am only thinking about kernel level features.]
- Very robust full featured statefull packet munging, filtering, notifiying thing (aka firewall).
- IPv6
- Support for 64bit address spaces and CPUs. (Where is the ia64 or x86-64 Windows on this?)
- NUMA (Does some version of Windows support Non-Unified Memory arch, may be something from Wang or some other dinosaur company).
- I am sure there are some esoteric network protocols linux supports natively. But I am not so impressed by that.
- Ether-switching (aka bridging; plus some stateful inspection).
This is from the top of my head. NFS is probably another, but MS has that LanManager file system, CIFS.
-- I am not a fanatic, I am a true believer.
In no particular order...
1 netfilter (firewalling)
2 hotloading/unloading device drivers
3 software raid
1. I know windows has some limited firewalling, but it is really a joke compared to iptables. SNAT/DNAT/TOS/Full customizable logging/Packet Mangling/ARP/TTL, I could go on. Many many commercial firewalls are based on linux, just look around.
2. I still have a windows box for my wife/kids. I used to have a scanner hooked up to it via an Adaptec 2904 SCSI adapter. Problem is windows would only "scan the bus" on boot, so if it wasn't turned on when I booted the machine (had to turn scanner off when not in use as the lamp stayed on all the time) I had to reboot windows.
Moved it to my Linux machine, now I just rmmod aic7xxx then modprobe aic7xxx every time I want the scanner. I even wrote a tiny little shell script and put an icon on the desktop so my wife/kids can do it.
#3 is a life saver to. At work we have a Linux backup server that has four 60 gig IDE hard drives in a raid5. The whole thing cost us less then $700 2 years ago for 180 gigs of file server/backup space. We are upgrading the drives to 120 gigs soon to double our space. Nothing like having 60 days of full backups a click away.
I could actually go on and on and on. It is astonishing how many usefull features are in the Linux kernel (not to mention the user-space stuff). I converted from windows three years ago now and I can't ever imagine going back. Windows is just so... limited.
apt-get install redhat please god - Me (take it easy, I love Debian)
Available in Windows
- Support for 64bit address spaces and CPUs. (Where is the ia64 or x86-64 Windows on this?)
Available in Windows - Itanic XP was available in a limited release and 64-bit 2003 Server can be bought pretty much off-the-shelf
- NUMA (Does some version of Windows support Non-Unified Memory arch, may be something from Wang or some other dinosaur company).
Available in Windows. From your comments, you appear unclear as to what NUMA is - it's not "something from Wang or some other dinosaur company", it's a way of architecting non-symmetrical MP hardware. Go read Greg Pfister's "In Search of Clusters", then go and discover how some Big Computers work - try Sequent (now part of IBM), Cray, SGI and Sun for details
- I am sure there are some esoteric network protocols linux supports natively. But I am not so impressed by that.
Yeah, Windows has those too...
Now, does Linux support hot-swap RAM? Hot-swap PCI cards? Useable async I/O, even (and no select(3) doesn't count...)? Does it even have a kernel debugger? :-)
Cheers
Jon
Just in case you don't know, he's a millionaire.
(More importantly though, he's involved in something he loves.)
"leave" is used in the old fashioned sense of "permission". So the phrase means "permitted absence". Contrast with military term AWOL=Absent WithOut Leave (ie, permission).
> I have had 2.4.20 completely freeze on me, couldn't even ssh in to it.
There are interrupt sharing issues with certain hardware in 2.4.20. I was able to get around it by swapping pci cards around. 2.4.21 was recently released and may even have a real fix. I haven't read the changelogs, let alone returned my hardware to its original configuration to check.
Contrast this to XP, which, with ACPI enabled, puts ALL devices on the same interrupt.