|Og dreams of kernels||Greg KH||2 years 29 weeks ago|
|Re: Old IPSEC bug||Theo de Raadt||2 years 13 weeks ago|
|Re: Allegations regarding OpenBSD IPSEC||Rod Whitworth||2 years 13 weeks ago|
|Re: Allegations regarding OpenBSD IPSEC||Jason L. Wright||2 years 14 weeks ago|
|Re: Allegations regarding OpenBSD IPSEC||Bob Beck||2 years 14 weeks ago|
|Allegations regarding OpenBSD IPSEC||Theo de Raadt||2 years 14 weeks ago|
Andrew Miklic recently announced progress on porting the Hurd gnumach microkernel to the Alpha processor. I approached him for more details on his efforts, and he explained:
"GNU Mach is the microkernel for the HURD. The rest of the HURD _should_ be platform-independent, so that the rest of the HURD/debian packages _should_ be functional on Alpha once the microkernel is up-and-running without requiring any more than a cross-compile.
"At this point, GNU Mach on Alpha is compiling, but not booting. To get this far was not that difficult for Alpha because some really old source code from the original CMU Mach 2.5 source was available, and whatever other code that exists for x86 but not on Alpha I just "stubbed" for the time being. To get it booting, I need to go back to the files that I "stubbed" and implement them properly."
Linus recently returned from a two week vacation, announcing the release of "a largish 2.5.8-pre1 patch". Following the announcement, he commented on the earlier April Fool's message. He says, "PS.
A recent thread on the lkml discussed the addition of another scheduler priority. Currently, referring to 'linux/include/schedule.h' in 2.4 (follows), there are three such priorities available. SCHED_FIFO, SCHED_RR and SCHED_OTHER. The additional priority in question is SCHED_IDLE.
This first two, SCHED_FIFO and SCHED_RR, are for real time processing. The former handles processes "first in, first out". Once a SCHED_FIFO process gets the CPU, it keeps it until finished, or until a higher priority process comes along. The latter handles processes "round robin", sharing time slices among all SCHED_RR process of equal priority. The third, SCHED_OTHER, is the standard, aka "time-share processing".
As for the discussed potential SCHED_IDLE, it would schedule a process to run only when the CPU was idle, not busy with other tasks. In the following thread, Robert Love says, "There is just a lot more to SCHED_IDLE than 'make the task only run when nothing else wants to'", explaining that as simple as it sounds, a lot of work is involved in the implementation.
Par for the course, "shocking" email hit the lkml on April 1'st attributed to Linus Torvalds. This year's email stated, "Linux needs new leadership", putting Linus' succesor up to a vote. A surprising number of people thought the email was real, expressing dismay at Linus' "choice" to abandon Linux. The contrived email follows.
OpenBSD founder Theo de Raadt announced today that OpenBSD now supports wavelan bridging: "In the past, many of you have asked if you can do wavelan bridging. We kept saying no, because you cannot run the wavelan in AP mode. There's no way to to put these cards into AP mode. For prism II based cards, however, you now can. Be sure to use a very new firmware on the cards, though."
Jeremy Jackson asked "which kernel debugger is 'best'?" on the Linux Kernel Mailing List.
The Enterprise Volume Managment System Project has released v1.0.0 of their storage management system. The GPL'd code is developed by IBM, and is compatible with both stable kernel 2.4.17 and development kernel 2.5.3. Find more information on their docs page, and in the latest release notes.
"The Enterprise Volume Management System ( EVMS ) Project has the goal of providing unparalleled flexibility and extensibility in managing storage. It represents a new approach to logical volume management, as the architecture introduces a plug-in model that allows for easy expansion or customization of various levels of volume management."
"We did indeed lose the primary disk (IBM 40GB, I am starting to lose all the respect I had for IBM drives, this is one of many that has failed on me personally). I have restored from the backup disk, and in the process redone hardlinks across all the linux kernel trees, which saved about 5GB (nice). All trees which are now on bkbits.net check clean, which means BK thinks all the files are there and that the checksums are correct, a fairly reasonable indication that we are in good shape."
Find an earlier discussion about BitKeeper here. Larry's full emails follow.
As originally submitted to KernelTrap by gncuster and reported on OSNews, the AtheOS operating system has been forked by Bill Hayden. The new creation is temporarily called New Atheos while Bill secures the domain names and trademark for the official and as-of-yet unnannounced name. Essentially he has merged the AtheOS and BeOS API's, porting it all to run on the Linux kernel. This results in significantly increased driver support, powerPC support, and the ability for most BeOS programs to compile and run with little or no changes. There has not yet been any source released, nor a date set for the official release.
Bill says, "I forked Atheos about 6 months ago and have been continuously developing it since that time. I've taken it in some very new directions. I should warn you that some of you will absolutely love the changes, and some of you will perhaps feel that the "dream" of Atheos has been sold out." Bill's announcement and much of the resulting thread follows. Find more information about AtheOS in this earlier review from OSNews.
Theo de Raadt recently announced that the Sparc64 port of OpenBSD has had a memory model change. Due to this change, there will be no supported upgrade from 3.0 to the upcoming 3.1 release. Instead, one will need to reinstall... In rather non-technical speak, Theo explains, "You must reinstall, due to the binaries having been changed in fiddly internal ways." He adds, "So just reinstall if you have a sparc64, ok? You will be happy. Some compiler bugs are fixed as a result!"
This only affects the Sparc64 port - no others.
Anton Altaparmakov recently annouced the 2.0.0 release of NTFS for the Linux 2.5.x kernel. This version is targetted for 2.5 inclusion, and is claimed to offer around a 20% speed gain over earlier NTFS drivers. You can browse the source code here. Full details follow.
From the Linux-NTFS FAQ:
"NTFS is an abbreviation for New Technology Filesystem. 'NT' because it was originally used in Windows NT and a filesystem is just how the computer stores files on disk. Different operating system, stores files in different ways. NTFS is used by Windows NT, 2000 and XP."
The latest LWN kernel report is out. Included is a discussion of the 2.4 VM patches from Andrea Arcangeli.
Linuxdevices.com has an article by Clark Williams (Red Hat) comparing the pre-emption patch, and the low latency patch. He includes performance graphs, and most interestingly, the performance benefits of using both patches. It's good to see that two different approaches can be combined to come up with even better results.