Tim Schmielau posted a patch that (finally?) fixes the ugly 497-day uptime problem, with minimal cruft. It seems like a pretty trivial patch (mainly dealing with types).
In a recent posting to the lkml, M. Edward Borasky asked on the current status of the 2.4 kernel's virtual memory system. He points out, "A few months ago, there was a flurry of reports from people having difficulties with memory management on large machines (ia32 over 4 GB). I've seen a lot of 2.4.x-yy kernels go by and much VM discussion, but what I'm *not* seeing is reports of either catastrophic behavior or its absence on large machines."
The resulting thread is an interesting read. William Lee Irwin III replied, "The catastrophic failures are still happening, in fact, the last lse-tech conference call a week or two ago was dedicated at least in part to them." He goes on to credit Andrea Arcangeli with the most visible efforts to fix problems in this arena. (Try Andrea's -aa kernel patch to see his latest work)
LinuxWorld recently posted an article by Richard Stallman (RMS) titled, Linux, GNU, and freedom. It begins as a reply to an earlier article by Joe Barr (The Stallman factor), explaining an incident with the SIGLINUX user group. RMS continues on to talk about the reason why the name 'GNU/Linux' is important, the influence Linus has, and his decision to utilize the BitKeeper tool.
Read on for more information, including a full mirror of the article.
Linus released the Linux development kernel 2.5.17 last night. He summarized the changes in two sentences, "Various FS updates (including merges of quota and iget_locked), and Makefile cleanups from Kai. And yet more TLB shootdown stuff."
This release includes the new and improved changelog format earlier story]. It combines some of the verbosity of BitKeeper changelogs with the readability of the old style kernel changelogs. Those watching the development kernel evolve will notice that Linus has been good to his word, releasing new kernels with increasing rapidity. Browse the latest development kernels on kernel.org.
The fourth annual Ottawa Linux Symposium will be happening on June 26th-29th in Ottawa, Canada. Dr. Stephen Tweedie of ext3 fame will be a keynote speaker. The website offers an impressive list of speakers, as well as numerous 'Bird of a Feather Sessions'.
Larry McVoy, the creator of BitKeeper, has also been invited, and offered to lead a discussion on the reason for BitKeeper's unique licensing. Richard Gooch, also speaking at the 2002 OLS, has urged him to speak, as Larry says, "I explained our reasons for why it is the way it is and he urged me to show up and explain it to a wider audience, I apparently explained things that weren't well known. I'm happy to do that, you can bring tomatoes :)".
I have spoken with Larry myself, and found his reasons to be quite sound. You can show up in Ottawa next month and decide for yourself.
OpenBSD 3.1 is out of the door! Many on the OpenBSD mailing lists have already reported receiving the new CDs. From Todd Miller's announcement email:
"It is our pleasure to officially announce the release of OpenBSD 3.1. This year OpenBSD turns 7 years old. In celebration of this milestone, we invite you to enjoy our 11th release on CD-ROM (and 12th via FTP)."
With this release, the OpenBSD home page has increased its security claim to, "Five years without a remote hole in the default install!"
FreeBSD 4.6 has entered the release candidate stage, destined for final release around June 1st. The complete FreeBSD release schedule can be found here. The quality assurance testing guide highlights three significant changes in this release: there is a major update for DHCP client support, the i4b SyncPPP driver has been merged with the main sppp driver, and XFree86 has been upgraded from version 3.3.6 to version 4.2.0. Check the release notes for the most up to date information. Though currently only available for i386, the Alpha version is also on its way.
Murray Stokely's 4.6-RC1 announcement email follows.
Update: (5/19/2) RC2 has been released to fix a sysinstall buffer truncation issue that affected the installtion of GNOME. Announcement below.
Keith Owens recently announced for the third time that kbuild 2.5 is ready for inclusion into the Linux 2.5 development kernel tree [earlier story]. Keith says, "Third and final attempt. Original sent on May 2, second mail sent on May 14, still no response from Linus." He goes on to explain, "I would like kbuild 2.5 to go in in the near future. Keeping up to date with kernel changes is a significant effort, Makefiles change all the time, especially when major subsystems like sound and usb are reorganised."
Many have offered support of the new build system. In the most recent thread, Robert Love said, "Fwiw, I like kbuild-2.5 and want it in." Dave Jones summarized his own support saying, "If kbuild2.5 is faster, and produces the same end result (or better still, more accurate builds), there's no valid reason to ignore it that I can see."
However, so far its seems that Linus has chosen to ignore Keith's emails. Perhaps there's something behind the scenes prompting this silence. Or perhaps it's just a matter of time.
GCC 3.1 has been officially released. Mark Mitchell sent out the announcement, in which he says,
"In this release, we focused more on quality than new features; many bugs were fixed. We worked very hard to fix bugs that were introduced in GCC 3.0, but that were not present in previous releases of the compiler. We also worked hard to eliminate new bugs."
When Linux creator Linus Torvalds began using the BitKeeper (BK) source control tool for managing the 2.5 Linux development kernel [earlier story], one of the big fears people put forward was that all Linux kernel developers would eventually be forced to use this tool. (The BK license is a major source of contention.
In late February we noted that the stable Linux kernel 2.2.21 could be expected around March 10th, 2002 [earlier story]. As it turns out, 2.2.21-rc4 was released today, about which Alan Cox says, "Unless something bad turns up this will be the final 2.2.21." There have been numerous fixes applied since late February. Details follow in Alan's changelog.